Strategies for Reforestation under Uncertain Future …ahamann/publications/pdfs/...Strategies for Reforestation under Uncertain Future Climates: Guidelines for Alberta, Canada Laura

Strategies for Reforestation under Uncertain FutureClimates: Guidelines for Alberta, CanadaLaura K. Gray*, Andreas Hamann

Department of Renewable Resources, University of Alberta, Edmonton, Alberta, Canada

Abstract

Background: Commercial forestry programs normally use locally collected seed for reforestation under the assumption thattree populations are optimally adapted to local environments. However, in western Canada this assumption is no longervalid because of climate trends that have occurred over the last several decades. The objective of this study is to show howwe can arrive at reforestation recommendations with alternative species and genotypes that are viable under a majority ofclimate change scenarios.

Methodology/Principal Findings: In a case study for commercially important tree species of Alberta, we use an ecosystem-based bioclimate envelope modeling approach for western North America to project habitat for locally adapted populationsof tree species using multi-model climate projections for the 2020s, 2050s and 2080s. We find that genotypes of species thatare adapted to drier climatic conditions will be the preferred planting stock over much of the boreal forest that iscommercially managed. Interestingly, no alternative species that are currently not present in Alberta can be recommendedwith any confidence. Finally, we observe large uncertainties in projections of suitable habitat that make reforestationplanning beyond the 2050s difficult for most species.

Conclusion/Significance: More than 50,000 hectares of forests are commercially planted every year in Alberta. Choosingalternative planting stock, suitable for expected future climates, could therefore offer an effective climate changeadaptation strategy at little additional cost. Habitat projections for locally adapted tree populations under observed climatechange conform well to projections for the 2020s, which suggests that it is a safe strategy to change current reforestationpractices and adapt to new climatic realities through assisted migration prescriptions.

Citation: Gray LK, Hamann A (2011) Strategies for Reforestation under Uncertain Future Climates: Guidelines for Alberta, Canada. PLoS ONE 6(8): e22977.doi:10.1371/journal.pone.0022977

Editor: Soo-Hyung Kim, University of Washington, United States of America

Received April 27, 2011; Accepted July 4, 2011; Published August 10, 2011

Copyright: � 2011 Gray, Hamann. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permitsunrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Funding: This study was supported by a Natural Sciences and Engineering Research Council of Canada (NSERC)/Industry Collaborative Development GrantCRDPJ 349100-06. Government funders included the NSERC and the Alberta Forest Research Institute. Industry co-sponsors included Alberta-Pacific ForestIndustries, Ainsworth Engineered Canada LP, Daishowa-Marubeni International Ltd., Western Boreal Aspen Corporation, and Weyerhaeuser Company Ltd.. Thefunders had no role in study design, data collection and analysis, decision to publish, or preparation of this manuscript.

Competing Interests: AH received a research grant and LKG received a graduate student stipend that was co-funded by the commercial companies listed in thefinancial disclosure. This does not alter the authors’ adherence to all the PLoS ONE policies on sharing data and materials.

* E-mail: [email protected]

Introduction

Reforestation with planting stock that is grown in nurseries is a

widely used practice in western Canada and elsewhere. Forest

companies and provincial agencies in Alberta plant approximately

80 million seedlings to reforest more than 50,000 hectares

annually. For successful reforestation programs, planting stock

must be both genetically well adapted to the target environment

and contain a sufficient amount of genetic diversity. Generally,

two decisions have to be made when selecting planting stock. First,

an appropriate species has to be chosen for a planting site. Usually,

forest sites can support several tree species, allowing forest

managers to choose which species best fit their economic or

ecological objectives. The second choice concerns the genetic

makeup of reforestation stock. Most widespread tree species show

adaptation of local populations to different macroclimatic

conditions that are frequently observed over latitudinal or

elevational gradients, e.g. [1]. To minimize the risk of maladap-

tation most jurisdictions legislate seed transfer guidelines or seed

zones, which restrict how far seed may be moved from a collection

location to a planting site [2,3]. Under the assumption that local

populations are optimally adapted to the environments in which

they occur, prescribing reforestation with species and genotypes

collected near the planting site can reduce the risk of maladap-

tation.

In Alberta, movement of seed is regulated with seed zones, a

system of approximately 60 geographic delineations for forested

areas of the province (Figure 1, map inset). These seed zones are a

subdivision of the Alberta Natural Regions and Subregions

ecological classification system [4]. Seed can be freely moved

within the seed zone or origin, but transferring seed outside seed

zone boundaries is usually prohibited [5]. Using fine scale

ecosystem classifications as a proxy for the genetic structure of

tree species is a common practice when lacking genetic

information. As genetic data become available from long-term

field experiments, fine scale seed zones are usually consolidated

into larger units if no genetic differentiation between adjacent

zones is found [2,3].

PLoS ONE | www.plosone.org 1 August 2011 | Volume 6 | Issue 8 | e22977

Although this system of governing seed movement has been

successfully used in many parts of the world, the key assumption

that local tree populations are optimally adapted to the

environments in which they occur, may no longer be valid. For

example, Alberta has experienced a warming trend of 0.8uC and a

decrease of about 10% in precipitation over the last 25 years [6].

Reciprocal transplant experiments have shown that there is now a

substantial mismatch between local populations and the environ-

ments in which they occur, leading to sub-optimal growth [7].

Furthermore, large-scale dieback of forest trees related to drought

stress has been observed along the southern edge of the boreal

forest [8,9,10]. The latter study estimates that drought-related

dieback of forest over the last decade has resulted in 45 Megatons

of dead biomass in central Alberta, representing 20% of the total

aboveground biomass.

Recognizing that management interventions are necessary to

maintain forest health and productivity in the face of climate

change, the Alberta government released interim seed transfer

guidelines in 2009, allowing upward and northward transfers

across adjacent seed zone boundaries within the natural subregion

of origin [11]. Larger distance seed transfers may be allowed but

require case-by-case approval from the Alberta Tree Improvement

and Seed Center [11]. We think that this policy framework can be

developed into an effective climate change adaptation strategy for

the forestry sector, and this study is meant to support decision

making by the provincial government of Alberta for selection of

species and genotypes that are well adapted to expected future

environments.

This study builds on a larger modeling effort that covers 15

commercially important forestry species of western North America

[12]. Here, we present a detailed regional analysis that can be used

to guide the reforestation activities in Alberta, and that may serve

as a template for other jurisdictions. We use multi-model

projections of species habitat for the 2020s, 2050s and 2080s to

aid species choice for reforestation. The goal is to arrive at species

recommendations that are viable under most climate change

scenarios. As a second step, we determine suitable genotypes for a

given planting site. Given the considerable uncertainty in climate

change projections, we provide multiple seed source recommen-

dations that approximately match expected future environments.

Multiple seed sources could be prescribed to enhance genetic

diversity in the landscape to hedge against uncertainty. We also

provide multiple choices of seed sources to allow flexible

implementation of assisted migration prescriptions in the face

logistical constraints in seed supply that forest companies and

provincial agencies face.

Methods

Climate envelope modelingThis study builds on an ecosystem-based modeling method

developed by Hamann & Wang [13] and Mbogga et al. [14]. The

approach characterizes the climate space of delineated ecosystem

polygons, which represent habitat for individual species popula-

tions. The ecosystem units are then predicted as a dependent class

variable using climate conditions under various future scenarios as

predictor. Predictions were performed with an ensemble classifi-

cation tree analysis implemented by the RandomForest software

package [15] for the R programming environment [16].

RandomForest grows multiple dichotomous decision trees from

bootstrap samples to predict a dependent class variable [17]. We

used 200 trees in this study, and the final predicted ecosystem was

determined by majority vote over all classification trees. As

dependent variable, we used the ‘‘seedzone’’ delineation of the

Natural Regions and Subregions of Alberta [4]. To determine

whether new species or seed sources from outside Alberta should

be introduced under climate change scenarios, we expanded the

model coverage to Canada and the United States west of 100ulongitude. Additional ecosystem units include the ‘‘variant’’ level

of the Biogeoclimatic Ecological Classification system for British

Figure 1. Climate of seed zones in Alberta, which are based on a hierarchical ecological classification system. Colors represent NaturalSubregions, and points in the scatterplot represent the finest units of forested ecosystems that govern seed transfer in reforestation. The delineationscorresponding to the scatterplot are shown on the map. The expected shift of a mean climate point for Alberta (1961–1990) representing the rangeof 18 climate change scenarios is indicated by ellipses (2020s, 2050s, 2080s).doi:10.1371/journal.pone.0022977.g001

Reforestation under Climate Change


Columbia [18]. For other Canadian provinces we used the

‘‘ecodistrict’’ level of the National Ecological Framework for

Canada [19], and for the United States we used the ‘‘level IV’’

classification of the Ecoregion System [20]. From each of these

ecosystem classes we randomly sampled 100 grid cells at 1 km

resolution, which we climatically characterized, and subsequently

used as training data for classification tree analysis.

Climate data and climate projectionsWe used interpolated climate data for the 1961–1990 normal

period, covering the United States and Canada west of 100ulongitude. Interpolation of weather station data was performed

with the Parameter Regression of Independent Slopes Model [21]

for monthly minimum temperature, maximum temperature and

precipitation. We enhanced this data with lapse-rate based down-

scaling to 1 km resolution and an estimation of biologically

relevant climate variables with a software package that is freely

available at http://www.ualberta.ca/,ahamann/climate.html

[6,22]. Ten predictor variables with low collinearity were chosen,

representing both seasonal and annual climate variables. This

includes mean annual temperature, mean warmest month

temperature, mean coldest month temperature, continentality

(difference between mean January and mean July temperature),

mean annual precipitation, growing season precipitation (May to

September), the number of frost free days, and the number of

growing degree days above 5uC. These variables are described in

more detail by Wang et al. [23]. We also included two dryness

indices: annual and summer climate-moisture index according to

Hogg [24].

To generate future climate projections for the 2020s, 2050s and

2080s we overlaid projections from general circulation models

expressed as the difference from the 1961–1990 normal using the

same software package as above. For each future period, climate

projections were based on four SRES emission and population

growth scenario families (A1FI, A2, B1, B2), implemented by five

modeling groups (CGCM, Canada; CSIRO2. Australia;

HADCM3, United Kingdom; ECHAM4, Europe; and PCM,

United States). Two model-emission scenario combinations

(ECHAM4-A1FI and ECHAM4-B1) were unavailable, resulting

in 18 climate projections per time period. Similar to GCM

projections, recent climate conditions can be expressed as

difference from the 1961–1990 normal period (also referred to

as anomaly). We use the 1997–2006 decadal anomaly to represent

observed climate change over a 25-year period (the midpoint of

the 1961–1990 climate baseline period and the midpoint of the

recent decadal average: 1975 to 2000).

Species projections and model validationWe use projected ecosystem units to represent populations of

tree species and to derive predictions of species habitat. The

frequency and probability of presence of major forest tree species

in ecosystem units was calculated from 54,716 forest inventory

plots covering western North America. This includes provincial

databases from British Columbia previously described in Hamann

et al. [25]. For Alberta we used permanent and temporary forest

inventory plots as well as the Ecological Site Information System

(ESIS) database provided by the Government of Alberta [26]. For

all the sample plots in western Canada, an estimated percent areal

cover of the canopy projected to the ground, scaled by the total

canopy of the forest inventory plot was used for species frequency.

In the western United States we rely on the Forest Inventory and

Analysis database [27], where we used the percent basal area was

used as a proxy for frequency because the percent areal cover of

the canopy was unavailable. Species frequency for each ecosystem

unit was calculated as the average across all sample plots that fall

within an ecosystem polygon. The probability of presence of a

species was simply calculated as the proportion of the inventory

plots within the ecosystem polygon where the species was present.

To assess the predictive accuracy of bioclimate envelope models

for individual species, we calculate the area under the curve (AUC)

of the receiver operating characteristics (ROC) curve of the

probability of species presence. The AUC value measures the

ability of the model to detect a species where it is known to be

present against its ability to correctly predict where the species is

known to be absent [28,29]. All ROC and AUC calculations were

carried out with the ROCR package [30] for the R programming

environment [16].

Five commercially important conifer tree species occur in

Alberta: black spruce (Picea mariana (Miller) Britton), Douglas-fir

(Pseudotsuga menziesii (Mirbel) Franco), lodgepole pine (Pinus contorta

Douglas ex Loudon), jack pine (Pinus banksiana Lambert, Descr.),

and white spruce (Picea glauca (Moench) Voss). Ponderosa pine

(Pinus ponderosa Douglas ex Lawson & C. Lawson) is projected to

gain suitable habitat in Alberta in the future [12] and was

therefore also included in this analysis. The scientific names are

according to the Flora of North America Editorial Committee

[31].

Seed source recommendationsMultiple options of seed sources for reforestation under current

and future climates were derived with a multivariate measure of

climate similarity. The objective was to find seed sources that best

match a target region under observed and projected climate

change. To quantify this match, we use the squared Mahalanobis

distance, calculated with the Ecodist package [32] for the R

programming environment [16]. Mahalanobis distances matrices

were calculated for 10 climate predictor variables described above,

and are reported for seed zone units characterized under current

climate and under ensemble scenarios for the 2020s, 2050s, and

2080s. The Mahalanobis distance is a normalized Euclidean

distance that weighs individual variables according to their

collinearity with all other variables [33]. Variables that are

perfectly correlated are weighted as a single variable in distance

calculations, while the Mahalanobis distance for completely

independent variables would equal the Euclidean distance. We

transformed all climate variables individually to conform to a

normal distribution before distance calculations. The Ecodist

package further transforms all variables into units of standard

deviations around a variable mean of zero prior distance

calculations, so that the weight of climate variables is independent

of their units of measurement.

Results

Alberta climatology and climate change projectionsThe climatology of Alberta’s ecological regions and seed zones

is primarily driven by a latitudinal temperature gradient, and

precipitation patterns that are related to the regional topography.

The Rocky Mountain Foothill and Montane ecosystems receive

the largest amounts of precipitation (500–700 mm) with mean

annual temperatures around 2uC (Fig. 1, blue shades). Note that

the outlying Montane ecosystem represents the Cypress Hill

region, a forest island in the southeast of the province’s grasslands

(yellow). Parklands (orange) represent a transitional zone between

grasslands and the boreal forest. Ecosystems of the boreal forest

(Fig. 1, green shades) span a diagonal from approximately 400 mm

precipitation and 24uC temperature to 500 mm precipitation and

2uC temperature. The diagonal arrangement of Natural Subre-



Table 1. Species statistics and model accuracy.

Global Statistics Alberta Statistics

Species Presence Samples1 Range Size (square km) AUC Presence Samples2 Range Size (square km) AUC

Black spruce 4,489 710,748 0.90 1,750 385,708 0.85

White spruce 7,115 848,866 0.88 3,606 438,013 0.79

Douglas-fir 8,808 1,002,592 0.88 269 9,952 0.91

Lodgepole pine 11,275 1,016,718 0.82 3,813 219,364 0.79

Ponderosa pine 3,967 591,394 0.88 0 0 NA

Jack pine 325 229,194 0.99 322 201,255 0.97

1Out of 54,716 sample plots, including non-forested plots.2Out of 16,391 sample plots, including non-forested plots.doi:10.1371/journal.pone.0022977.t001

Figure 2. Seed zones projections and consensus of habitat maintenance under projected climate change for white spruce inAlberta. Colors represent broad seed sources corresponding to Natural Subregions (upper row), and the gray scale represents the consensus thathabitat is maintained for white spruce for 18 climate change scenarios for the 2020s, 2050s, 2080s (lower row). We require at least a 70% probabilitythat habitat is maintained to make a seed source recommendation.doi:10.1371/journal.pone.0022977.g002



gion classes (shades of green) suggests that the precipitation/

evaporation balance distinguishes these major ecosystem classes.

To visualize projected climate change relative to the 1961–1990

normal climatology, we added the current climatology and

projections for a central boreal forest location, an area centered

around 56u latitude and 115u longitude (Fig. 1, open circle). The

range of uncertainty in predicted temperature and precipitation

values is represented by ellipses. The range of projected climate

change varies for different locations in Alberta and cannot be

comprehensively visualized in this plot. It is clear, however, that

the uncertainty in climate change projections stands in strong

contrast to the precision with which reforestation is managed

trough seed zones at present (each point in Fig. 1 represents a

separate seed zone). Even for the 2020s, similar ellipses drawn at

other locations may easily encompass several seed zones as

possible alternatives for obtaining reforestation material under

climate change. At least in this simple, two-dimensional visuali-

zation, it appears challenging to pinpoint seed zone recommen-

dations for the 2050s and 2080s, where similar ellipses drawn at

various locations may regularly span several ecological subregions,

indicated by different colors in Fig. 1.

Projections of tree species habitatArea Under the Curve (AUC) statistics suggest that the

predictive accuracy of the ecosystem-based climate envelope

model for Alberta is satisfactory (Table 1). Local AUC statistics

for Alberta are similar to those for the global species range

predictions. In general terms, AUC values above 0.9 indicate

excellent predictive accuracy and AUC values above 0.8 indicate

good accuracy. An AUC value of 0.8 means that 80% of the time a

random sample from presence predictions will have a score greater

than a random selection from absence predictions across all

available probability thresholds to define a presence prediction. An

AUC value of 0.5 therefore indicates a random predictor and

values between 0.5 and 0.6 are generally considered a failed model

[29].

Habitat projections under future climate change scenarios are

shown in Fig. 2 for white spruce. Projections for other important

forestry species in Alberta are provided as Fig. S1 (black spruce),

Fig. S2 (Douglas-fir), Fig. S3 (lodgepole pine), Fig. S4 (jack pine).

In these figures, the black-and-white maps represent the consensus

of projections for 18 climate change scenarios. Black indicates that

all models agree that climate conditions will be suitable for a

species, and white indicates that all models agree that suitable

habitat is not available under any scenario. Grey shades represent

varying levels of uncertainty in future habitat availability. The

results for white spruce are numerically summarized in Table 2,

where habitat suitability is provided for selected seed zones of

Alberta (complete tables for all species are provided as Tables S1,

S2, S3, S4, S5).

Table 2. Suitable white spruce habitat expressed as % area ofseed zone for observed climate, and expressed as probabilityof habitat maintenance under climate change projectionsfrom 18 general circulation models.

White spruce Observed climate Projected climate

seed zones1 1961–1990 1997–2006 2020s 2050s 2080s

CM 1.1 100% 100% 100% 98% 75%

CM 1.2 100% 100% 100% 92% 67%

CM 1.3 100% 100% 100% 98% 71%

DM 1.1 100% 100% 99% 85% 56%

DM 1.2 99% 98% 88% 66% 50%

DM 1.3 100% 100% 74% 74% 59%

DM 2.1 73% 95% 74% 88% 57%

DM 2.2 99% 99% 67% 87% 69%

1A complete table for all white spruce seed zones is provided as Table S5.doi:10.1371/journal.pone.0022977.t002

Figure 3. Suitable habitat under projected under climate change for ponderosa pine in Alberta. There is large uncertainty whether thisspecies may become a viable forestry species in Alberta, with extensive areas of suitable habitat projected under some climate change scenarios, andvirtually no habitat under other climate change projection.doi:10.1371/journal.pone.0022977.g003



For white spruce (Fig. 2, Table 2), habitat is generally well

maintained into the future except for some of the current Dry

Mixedwood and transitional Parkland ecosystems. The ecosystem-

based habitat projections also convey where appropriate seed

sources for expected future climates may be found. For white

spruce we observe that seed sources adapted to drier and warmer

conditions (Parkland, Dry Mixedwood) should be suitable for an

increasing land base in Alberta in the future. In contrast, black

spruce is predicted to lose much of its climatically suitable habitat

in Alberta, especially in low elevation regions (Fig. S1, Table S1).

Douglas-fir is only a commercially viable forestry species in

Montane ecosystems in the southeast corner of the province.

However, habitat projections for Douglas-fir come with large

uncertainties (Fig. S2, Table S2). Climate scenarios that project

substantially increased temperature and precipitation for south-

western Alberta, such as the CGCM-A1F1 scenario, result in

largely extended habitat for Douglas-fir throughout the Foothill

ecosystems of Alberta. On average, however, suitable habitat

remains constant or is slightly reduced. The current distribution of

lodgepole pine in the foothills of Alberta appears to be well

maintained with reasonable certainty (Fig. S3, Table S3). Lastly,

habitat for jack pine, currently concentrated at lower elevations in

the northeast of the province, is predicted to rapidly decline under

most climate change scenarios (Fig. S4, Table S4).

Notably, no alternative species that are currently not present in

Alberta can be recommended with confidence, meaning that

suitable habitat is predicted under a clear majority of climate

change scenarios. Ponderosa pine (Fig. 3, Table 3) comes closest in

gaining habitat with sufficient confidence across multiple climate

change scenarios. By the 2050s, the most southern Montane

ecosystems of Alberta may become suitable according to

approximately half the 18 climate change scenarios we used.

Projections of appropriate seed sourcesIf habitat for a species is maintained under at least 70% of the

climate change scenarios, we also provide projections of suitable

seed sources. These projections are visualized in the series of color

maps in Fig. 2 and Figures S1, S2, S3, S4. In these figures, the

colors represent the broad Natural Subregions rather than

individual seed zones for the purpose of better visualizing shifts

in climate habitat. For white spruce, it is apparent that much of the

land base of Alberta will require reforestation stock that is adapted

to the warmer and drier ecosystems of the current Dry Mixed-

wood and Parkland ecosystems. In Table 4, more detailed

information is provided for individual seed zones. This table

provides alternative seed sources according to the climate match

under current and expected future climates. For example, by the

2020s the Central Mixedwood seedzone CM 1.1 is predicted to

closely match current Dry Mixedwood climate of the seed zone

DM 1.1, or the more southern Central Mixedwood seed zones

CM 1.2 and CM 1.3. These seed zones are also close matches

under observed climate change, represented by the 1997–2006

average climate, and might therefore be recommended as source

for planting material under a climate change adaptation strategy.

Complete tables for all seed zones and up to 10 alternative choices

are provided as Tables S6, S7, S8, S9, S10. Locations of

recommended seed choices originating outside of Alberta are

provided as Table S11.

Discussion

Species choice for reforestationTo minimize the probability of plantation failure in the face of

uncertain future climates, we think that the best strategy is to

ensure that species habitat is maintained under a wide range of

potential climate change scenarios. In this study we restrict our

Table 4. Seed zones with the best climate match, as measured with the multivariate Mahalanobis distances given in parenthesis.

Observed Climate Projected Climate

Seedzones1 1961–1990 1997–2006 2020s 2050s 2080s

CM 1.1 CM11(0), PAD11(0),AP11(0.1), CM13(0.6)

CM12(3.6), DM11(4),CM21(4.8)

DM11(1.6), CM12(2.1),CM13(2.1), PAD11(2.2)

DM11(3.9), CM31(4.8) [MT]42i(5.2)

CM 1.2 CM12(0), CM21(0.4),DM11(0.4), CM22(1.2)

CM12(3.2), CM24(3.2),CM21(3.6), CM23(3.6)

DM11(1.5), CM12(2),CM21(2.3), CM31(2.4)

CM31(3), DM21(3.3),DM12(4.1), CM32(4.3)

[MT]42i(3.8), 42k(4.2),DM21(4.3), CM32(5.2)

DM 1.2 DM12(0), DM13(0.8),LBH16(0.8), PRP11(0.8)

DM12(2.3), PRP11(2.6),DM13(2.6), CM31(3.6)

PRP11(0.4), DM13(0.6),DM12(1), DM21(1.3)

PRP11(1.2), DM21(1.4),CP11(1.5), DM13(1.5)

CP11(1.7), CP12(1.9),NF11(2), DM22(2.3)

DM 1.3 DM13(0), PRP11(0.3),CM33(0.6), DM12(0.8)

PRP11(2), DM13(2),NF11(2.3), MG11(2.3)

DM13(0.6), CM34(0.9),PRP11(0.9), DM21(1)

CP11(1.3), CP12(1.5),DM22(1.5), CM34(1.6)

DM22(2.2), CP11(2.3),CP12(2.7), CM34(2.9)

Recommendations for U.S. seed sources are preceded by their state of origin.1Complete tables for all seed zones with up to 10 alternative options is provided in Tables S6, S7, S8, S9, S10.doi:10.1371/journal.pone.0022977.t004

Table 3. Suitable ponderosa pine habitat expressed as % areaof seed zone for observed climate, and expressed asprobability of habitat maintenance under climate changeprojections from 18 general circulation models.

Ponderosa pine Observed Climate Projected Climate

Seed zones 1961–1990 1997–2006 2020s 2050s 2080s

CM 3.5 0% 0% 0% 11% 43%

DM 2.3 0% 3% 0% 40% 66%

LF 2.3 0% 3% 1% 42% 67%

M 1.1 0% 0% 7% 33% 30%

M 2.1 0% 0% 0% 13% 58%

M 2.2 0% 1% 2% 19% 44%

M 3.2 0% 1% 0% 11% 36%

M 4.3 0% 0% 4% 30% 46%

M 4.4 0% 0% 4% 49% 67%

M 4.5 0% 25% 19% 54% 51%

M 5.6 0% 1% 10% 46% 60%

doi:10.1371/journal.pone.0022977.t003



reporting to a threshold of at least 70% of the models to agreeing

that species habitat will be maintained. Practitioners may want to

set higher thresholds for implementing large-scale reforestation

programs to minimize risks of plantation failure. On the other

hand, it should be noted that predicted loss of habitat does not

necessarily mean dieback or failure to reproduce for tree species.

Like most species distribution models, our approach predicts the

realized niche (that is the climate space where the species is found

to occur naturally) and not the larger fundamental niche space

(namely, all climate conditions that a species can tolerate).

By their nature, the predictions of the realized niche space are

more conservative as they account for biotic interactions. For

example, a tree species may be predicted to lose habitat because it

will be out-competed by other species that are better adapted to

the predicted environment. However, in a planting environment

with site preparation, controlled spacing, and removal of

competing vegetation, natural competition would be limited.

Secondly, the realized niche of trees may be determined by the

ability of seedlings to germinate under favorable conditions and

saplings to get established. Mature trees that have access to water

through a large root system tend to have a much larger

fundamental niche space than their offspring. Again, forest

managers can literally ‘‘push the envelope’’ of where a tree

species can be successfully grown by cultural treatments, such as

planting sturdy seedlings that were grown to a relatively large size

in a forest nursery.

Biotic interactions that are implicitly included in realized niche

models also include insect pests and diseases. A tree species might

be excluded from an area not because the environmental

conditions are unfavorable, but because the abiotic conditions

are also favorable for a forest pest to which the species is

susceptible. This mechanism might be particularly relevant to this

study area, as many insects and diseases are excluded from boreal

environments due to extreme cold in winter [34]. Species choice in

large-scale reforestation programs should be determined by the

maintenance of the realized niche under most climate change

scenarios, avoiding potential exposure of forest trees to pests and

diseases under a continued warming trend (Tables S1, S2, S3, S4,

S5 describe where the realized niche space is maintained).

Choice of genotypes for reforestationMatching genotypes to abiotic environments with the precision

of Alberta’s current system of seed zones is unlikely to be a sensible

strategy in the face of uncertain future climates. In fact, the current

level of precision may not even be necessary under constant

climate conditions. Forest trees are normally adapted to broad

environmental gradients with substantial within-population genet-

ic diversity [35]. Recent data from genetic provenance experi-

ments suggests that genetic differentiation of tree populations in

Alberta would occur at a much broader scale than the current seed

zone delineations [3,36,37]. As such data from long-term trials

become available for more species, general seed zones could be

consolidated into larger units to ease the administrative and

logistical burden of maintaining many separate seed collections for

reforestation needs. For this decision process, which should

synthesize genetic differentiation of tree populations, topo-edaphic

characteristics of seed zones, and climatic information, we

contribute a matrix of climatic similarity for current seed zones

in Table S6.

For the development of reforestation strategies under climate

change, we encourage practitioners to consult Tables S6 and S7,

which provide multiple choices of appropriate seed sources for

climate conditions observed over a recent decade and projections

for the 2020s. Ideally, seed sources should be used that appear as

options under the 1961–1990 reference climate, under 1997–2006

climate, and under 2020s climate projections. Several, consistently

suitable choices can usually be found. Making recommendations

for the 2050s and 2080s becomes difficult because of the large

uncertainties associated with climate projections in the more

distant future. We propose that this information might be used for

long-term planning, but not for guidance of seed sources in the

near future. Planting trees for 2050s and 2080s climate is not

sensible as seedlings will likely not survive current planting

environments. Also, we ultimately do not need to adapt to a

‘‘median climate change scenario’’ but to climate trends that

eventually materialize in Alberta. At this point, we do not know

with any reasonable amount of certainty what those conditions will

be by the end of the century.

In choosing seed sources for the immediate future, we should

further discuss the meaning of the Mahalanobis distances provided

in Tables S6, S7, S8, S9, S10. The values provide a measure of

climatic similarity (smaller = more similar) between seed zones

under 1961–1990 reference climate and future climate conditions

expected for these seed zones. The measure does not have an

interpretable dimension, and a larger distance does not necessarily

imply maladaptation of tree populations. Although this could be

the case, it should be noted that we do not have biological and

genetic data that demonstrates reduced fitness or productivity as a

function of any particular climate variable that is used for the

Mahalanobis distance calculation. Nevertheless, the alternate

choices provided in Tables S6, S7, S8, S9, S10 could still be

used to develop a simple portfolio strategy of adaptation to climate

change, where multiple seed sources that approximately match

current and 2020s climate are prescribed for reforestation. Such a

portfolio approach could continue to use the current seed zone

delineations as target areas. Use of multiple seed sources should

also include a mechanism for tracking reforestation success,

growth, and forest health of plantations to allow recursive

improvements [38].

Finally, we should note that importing seed and species from

other jurisdictions does not promise to be an important element of

a climate change adaptation strategy for the forestry sector in

Alberta. Only in small areas of the southern Rocky Mountain

Montane and Foothill ecosystem, habitat is projected to be suited

to populations originating in montane ecosystems of British

Columbia, and the dry conifer forests in Montana, South Dakota,

and Wyoming (Table S11). Of approximately 50 western North

American tree species that we investigated in a larger modeling

effort, no alternative species that are currently not present in

Alberta can be recommended with any confidence for reforesta-

tion under projected climate change.

Supporting Information

Figure S1 Seed zones projections and consensus ofhabitat maintenance under projected climate changefor black spruce in Alberta. Colors represent broad seed

sources corresponding to Natural Subregions (upper row), and the

gray scale represents the consensus that habitat is maintained for

black spruce under 18 climate change scenarios for the 2020s,

2050s, 2080s (lower row). We require at least a 70% probability

that habitat is maintained to make a seed source recommendation.

(PDF)

Figure S2 Seed zones projections and consensus ofhabitat maintenance under projected climate changefor Douglas-fir in Alberta. Colors represent broad seed





Douglas-fir under 18 climate change scenarios for the 2020s,

2050s, 2080s (lower row). We require at least a 70% probability

that habitat is maintained to make a seed source recommendation.

(PDF)

Figure S3 Seed zones projections and consensus ofhabitat maintenance under projected climate changefor logepole pine in Alberta. Colors represent broad seed



lofgepole under 18 climate change scenarios for the 2020s, 2050s,

2080s (lower row). We require at least a 70% probability that

habitat is maintained to make a seed source recommendation.

(PDF)

Figure S4 Seed zones projections and consensus ofhabitat maintenance under projected climate changefor jack pine in Alberta. Colors represent broad seed sources

corresponding to Natural Subregions (upper row), and the gray

scale represents the consensus that habitat is maintained for jack

pine under 18 climate change scenarios for the 2020s, 2050s,

2080s (lower row). We require at least a 70% probability that

habitat is maintained to make a seed source recommendation.

(PDF)

Table S1 Suitable black spruce habitat expressed as %area of seed zone for observed climate, and expressed asprobability of habitat maintenance under climatechange projections from 18 general circulation models.

(PDF)

Table S2 Suitable Douglas-fir habitat expressed as %area of seed zone for observed climate, and expressed asprobability of habitat maintenance under climatechange projections from 18 general circulation models.

(PDF)

Table S3 Suitable lodgepole pine habitat expressed as% area of seed zone for observed climate, and expressedas probability of habitat maintenance under climatechange projections from 18 general circulation models.

(PDF)

Table S4 Suitable jack pine habitat expressed as % areaof seed zone for observed climate, and expressed asprobability of habitat maintenance under climatechange projections from 18 general circulation models.

(PDF)

Table S5 Suitable white spruce habitat expressed as %area of seed zone for observed climate, and expressed asprobability of habitat maintenance under climatechange projections from 18 general circulation models.(PDF)

Table S6 Table of best matching seed sources for 1961–1990 climate. The multivariate Mahalanobis climate distance is

given in parenthesis.

(PDF)

Table S7 Table of best matching seed sources for 1997–2006 climate. The multivariate Mahalanobis climate distance is


(PDF)

Table S8 Table of best matching seed sources for 2020sclimate. The multivariate Mahalanobis climate distance is given

in parenthesis.

(PDF)

Table S9 Table of best matching seed sources for 2050sclimate. The multivariate Mahalanobis climate distance is given

in parenthesis.

(PDF)

Table S10 Table of best matching seed sources for2080s climate. The multivariate Mahalanobis climate distance is


(PDF)

Table S11 Locations of recommended seed choiceswhich originate outside of Alberta. For British Columbia

we report the relevant ecological ‘‘variants’’ and ‘‘zones’’ [18], and

for the United States we report the corresponding state and ‘‘level

III & IV’’ ecoregions [20].

(PDF)

Acknowledgments

For provision of databases and help with data preparation we thank Todd

Schroeder from the United States Forest Service, and Deogratias

Rweyongeza, Leonard Bernhardt and Ken Greenway from Alberta

Sustainable Resource Development. In addition, we thank Xianli Wang

and David Roberts for help with data preparation and analysis.

Author Contributions

Conceived and designed the experiments: LKG AH. Performed the

experiments: LKG AH. Analyzed the data: LKG AH. Contributed

reagents/materials/analysis tools: AH. Wrote the paper: LKG AH.

References

1. Morgenstern E (1996) Geographic Variation in Forest Trees. Genetic Basis and

Application of Knowledge in Silviculture. VancouverBC, , Canada: University

of British Columbia Press. 208 p.

2. Ying CC, Yanchuk AD (2006) The development of British Columbia’s tree seed

transfer guidelines: purpose, concept, methodology, and implementation. Forest

Ecology and Management 227: 1–13.

3. Hamann A, Gylander T, Chen P (2011) Developing seed zones and transfer

guidelines with multivariate regression trees. Tree Genetics & Genomes 7:

399–408.

4. NRC (2006) Natural regions and Subregions of Alberta. Natural Regions

Committee, Government of Alberta, Alberta Environment, Edmonton, Alberta,

ISBN 0-7785-4572-5.

5. SRD (2005) Standards for Tree Improvement in Alberta (STIA). Edmonton,

Alberta: Alberta Sustainable Resource Development, ISBN 0-7785-4082-0.

6. Mbogga MS, Hamann A, Wang TL (2009) Historical and projected climate data

for natural resource management in western Canada. Agricultural and Forest

Meteorology 149: 881–890.

7. Gray LK, Gylander T, Mbogga M, Chen P, Hamann A (2011) Assisted

migration to address climate change: recommendations for aspen reforestation

in western Canada. Ecological Applications 21: 1591–1603.

8. Hogg EH, Brandt JP, Kochtubajda B (2002) Growth and dieback of aspen

forests in northwestern Alberta, Canada, in relation to climate and insects.

Canadian Journal of Forest Research 32: 823–832.

9. Hogg EH, Brandt JP, Michaellian M (2008) Impacts of a regional drought on the

productivity, dieback, and biomass of western Canadian aspen forests. Canadian

Journal of Forest Research 38: 1373–1384.

10. Michaelian M, Hogg EH, Hall RJ, Arsenault E (2010) Massive mortality of

aspen following severe drought along the southern edge of the Canadian boreal

forest. Global Change Biology 17: 2084–2094.

11. SRD (2009) Alberta Forest Genetic Resource Management and Conservation

Standards (FGRMS). Edmonton, Alberta: Alberta Sustainable Resource

Development, ISBN 978-0-7785-8467-4.

12. Gray LK (2011) Assisted migration to address climate change: recommendations

for reforestation in western Canada. PhD Thesis, University of Alberta,

Edmonton, Alberta. 229 p.

13. Hamann A, Wang TL (2006) Potential effects of climate change on ecosystem

and tree species distribution in British Columbia. Ecology 87: 2773–2786.

14. Mbogga MS, Wang XL, Hamann A (2010) Bioclimate envelope model

predictions for natural resource management: dealing with uncertainty. Journal

of Applied Ecology 47: 731–740.



15. Breiman L (2001) Random forests. Machine Learning 45: 5–32.

16. R Development Core Team (2008) R: A language and environment forstatistical computing. Vienna, Austria: R Foundation for Statistical Computing,

ISBN 3-900051-07-0.

17. Cutler DR, Edwards TC, Beard KH, Cutler A, Hess KT (2007) Random forestsfor classification in ecology. Ecology 88: 2783–2792.

18. Meidinger D, Pojar J (1991) Ecosystems of British Columbia. Special ReportSeries 6. Research Branch, BC Ministry of Forests and Ranges, Victoria, British

Columbia, ISBN 0843-6452.

19. Selby CJ, Santry MJ (1996) A National Ecological Framework for Canada: Datamodel, Database and Programs. Ottawa, Ontario: Centre for Land and

Biological Resources Research, Research Branch, Agriculture and Agri-FoodCanada and State of the Environment Directorate, Environment Canada, ISBN

0-662-24107-X.20. EPA (2007) Ecoregion Maps and GIS Resources. Corvallis, OR: U.S.

Environmental Protection Agency, Western Ecology Division official website,

Available online at: http://www.epa.gov/wed, accessed 10 May 2008.21. Daly C, Halbleib M, Smith JI, Gibson WP, Doggett MK, et al. (2008)

Physiographically sensitive mapping of climatological temperature and precip-itation across the conterminous United States. International Journal of

Climatology 28: 2031–2064.

22. Mbogga M, Hansen C, Wang T, Hamann A (2010) A Comprehensive Set ofInterpolated Climate Data for Alberta. Government of Alberta, Publication

Number: Ref. T/235. ISBN: 978-0-7785-9184-9 (on-line edition), 978-0-7785-9183-2 (print edition).

23. Wang T, Hamann A, Spittlehouse DL, Aitken SN (2006) Development of scale-free climate data for western Canada for use in resource management.

International Journal of Climatology 26: 383–397.

24. Hogg EH (1997) Temporal scaling of moisture and the forest-grasslandboundary in western Canada. Agricultural and Forest Meteorology 84: 115–122.

25. Hamann A, Smets P, Yanchuk AD, Aitken S, N (2005) An ecogeographicframework for in situ conservation of forest trees in British Columbia. Canadian

Journal of Forest Research 35: (2553–2561).

26. Government of Alberta (2008) Ecological Site Information System (ESIS) electronic

database. Available: http://www.srd.alberta.ca/MapsFormsPublications/Maps/ResourceDataProductCatalogue/Ecological.aspx. Accessed 2008 May 10.

27. Bechtold WA, Patterson PL (2005) The enhanced Forest Inventory and Analysis

program - a national sampling design and estimation procedures. Ashville, NC:USDA Forest Service General Technical Report, SRS-80.

28. Fawcett T (2006) An introduction to ROC analysis. Pattern Recognition Letters27: 861–874.

29. Fielding AH, Bell JF (1997) A review of methods for the assessment of prediction

errors in conservation presence/absence models. Environmental Conservation24: 38–49.

30. Sing T, Sander O, Beerenwinkel N, Lengauer T (2005) ROCR: visualizingclassifier performance in R. Bioinformatics 21: 3940–3941.

31. Flora of North America Editorial Committee (eds) (1993+) Flora of NorthAmerica of Mexico. 15 vols., New York and Oxford.

32. Goslee SC, Urban DL (2007) The ecodist package for dissimilarity-based

analysis of ecological data. Journal of Statistical Software 22: 1–19.33. Mahalanobis PC (1936) On the generalised distance in statistics. Proceedings of

the National Institute of Science of India 12: 49–55.34. Volney WJA, Fleming RA (2000) Climate change and impacts of boreal forest

insects. Agriculture Ecosystems & Environment 82: 283–294.

35. Hamrick JL (2004) Response of forest trees to global environmental changes.Forest Ecology and Management 197: 323–335.

36. Rweyongeza DM, Dhir NK, Barnhardt LK, Hansen C, Yang RC (2007)Population differentiation of the lodgepole pine (Pinus contorta) and jack pine

(Pinus banksiana) complex in Alberta: growth, survival, and responses to climate.Canadian Journal of Botany 85: 545–556.

37. Rweyongeza DM, Barnhardt LK, Dhir NK, Hansen C (2010) Population

Differentiation and Climatic Adaptation for Growth Potential of White Spruce(Picea glauca) in Alberta, Canada. Silvae Genetica 59: 158–169.

38. Millar CI, Stephenson NL, Stephens SL (2007) Climate change and forests ofthe future: Managing in the face of uncertainty. Ecological Applications 17:

2145–2151.



Figure S1. Seed zones projections and consensus of habitat maintenance under projected climate change for black spruce in Alberta. Colors represent broad seed sources corresponding to Natural Subregions (upper row), and the gray scale represents the consensus that habitat is maintained for black spruce under 18 climate change scenarios for the 2020s, 2050s, 2080s (lower row). We require at least a 70% probability that habitat is maintained to make a seed source recommendation.

Figure S2. Seed zones projections and consensus of habitat maintenance under projected climate change for Douglas-fir in Alberta. Colors represent broad seed sources corresponding to Natural Subregions (upper row), and the gray scale represents the consensus that habitat is maintained for Douglas-fir under 18 climate change scenarios for the 2020s, 2050s, 2080s (lower row). We require at least a 70% probability that habitat is maintained to make a seed source recommendation.

Figure S3. Seed zones projections and consensus of habitat maintenance under projected climate change for logepole pine in Alberta. Colors represent broad seed sources corresponding to Natural Subregions (upper row), and the gray scale represents the consensus that habitat is maintained for lofgepole under 18 climate change scenarios for the 2020s, 2050s, 2080s (lower row). We require at least a 70% probability that habitat is maintained to make a seed source recommendation.

Figure S4. Seed zones projections and consensus of habitat maintenance under projected climate change for jack pine in Alberta. Colors represent broad seed sources corresponding to Natural Subregions (upper row), and the gray scale represents the consensus that habitat is maintained for jack pine under 18 climate change scenarios for the 2020s, 2050s, 2080s (lower row). We require at least a 70% probability that habitat is maintained to make a seed source recommendation.

Black spruce Observed Climate Projected Climate

seedzones* 1961-1990 1997-2006 2020s 2050s 2080s

BSA 1.1 98% 77% 96% 98% 89%BSA 1.2 100% 100% 100% 100% 90%CM 1.1 100% 100% 100% 94% 52%CM 1.2 100% 100% 100% 75% 34%CM 1.3 100% 100% 100% 95% 46%CM 2.1 100% 99% 98% 58% 29%CM 2.2 100% 100% 97% 60% 30%CM 2.3 100% 100% 97% 47% 22%CM 2.4 100% 100% 87% 52% 30%CM 3.1 99% 82% 59% 29% 17%CM 3.2 100% 64% 75% 42% 23%CM 3.3 100% 100% 96% 56% 33%CM 3.4 100% 100% 76% 36% 27%CM 3.5 68% 99% 15% 38% 21%DM 1.1 100% 100% 99% 62% 29%DM 1.2 99% 83% 73% 22% 7%DM 1.3 100% 95% 40% 11% 4%DM 2.2 99% 26% 24% 19% 14%DM 2.3 100% 87% 62% 28% 8%LBH 1.1 100% 100% 100% 98% 60%LBH 1.2 98% 88% 100% 100% 76%LBH 1.3 100% 94% 100% 97% 61%LBH 1.4 100% 100% 100% 77% 40%LBH 1.5 100% 100% 99% 71% 45%LBH 1.6 100% 100% 98% 70% 39%LBH 2.1 100% 100% 100% 100% 83%LF 1.1 95% 100% 100% 94% 65%LF 1.2 95% 100% 82% 25% 20%LF 1.3 100% 81% 84% 77% 65%LF 1.4 100% 57% 88% 78% 59%LF 1.5 100% 99% 43% 57% 40%LF 2.1 100% 67% 83% 77% 57%LF 2.2 100% 87% 74% 61% 32%LF 2.3 81% 21% 72% 37% 14%NM 1.1 100% 100% 100% 100% 71%NM 2.1 99% 100% 100% 100% 81%UBH 1.1 99% 97% 99% 100% 77%UBH 1.2 100% 100% 100% 95% 62%UBH 1.3 94% 100% 100% 91% 61%UF 1.1 100% 81% 100% 93% 58%UF 1.2 100% 57% 100% 95% 77%UF 1.3 93% 36% 98% 79% 44%UF 1.4 98% 72% 99% 88% 60%UF 1.5 77% 63% 78% 68% 29%UF 2.4 100% 99% 96% 93% 67%UF 2.5 65% 90% 86% 86% 53%

Table S1. Suitable habitat expressed as % area of seed zone for observed climate, and expressed as probability of habitat maintenance under climate change projections from 18 general circulation models.

Douglas-fir Observed Climate Projected Climate

seedzones* 1961-1990 1997-2006 2020s 2050s 2080s

M 2.2 91% 70% 55% 57% 64%M 4.3 97% 56% 39% 53% 50%M 4.5 100% 100% 78% 73% 51%M 5.3 77% 38% 49% 50% 57%M 5.5 100% 100% 88% 71% 56%M 5.6 85% 87% 85% 78% 72%


Lodgepole Pine Observed Climate Projected Climate

seedzones* 1961-1990 1997-2006 2020s 2050s 2080s

BSA 1.1 98% 59% 67% 61% 47%CM 3.2 48% 15% 70% 55% 45%CM 3.3 21% 71% 74% 68% 58%CM 3.4 100% 86% 87% 72% 69%CM 3.5 95% 7% 94% 76% 49%LBH 1.2 96% 20% 46% 34% 17%LBH 1.5 99% 44% 42% 40% 24%LBH 2.1 100% 89% 65% 49% 24%LF 1.2 100% 100% 88% 59% 66%LF 1.3 100% 75% 100% 99% 87%LF 1.4 100% 96% 99% 98% 85%LF 1.5 100% 54% 99% 91% 69%LF 2.1 100% 99% 98% 98% 88%LF 2.2 100% 99% 89% 86% 61%LF 2.3 100% 100% 65% 51% 51%M 1.1 99% 75% 27% 23% 11%M 2.1 100% 100% 99% 93% 67%M 2.2 100% 89% 70% 63% 61%M 3.2 100% 98% 79% 80% 76%M 4.1 100% 100% 86% 85% 78%M 4.2 100% 100% 99% 87% 91%M 4.3 100% 100% 55% 57% 51%M 4.4 100% 100% 60% 58% 48%M 4.5 98% 98% 78% 73% 51%M 5.1 100% 95% 85% 80% 79%M 5.3 100% 100% 89% 81% 71%M 5.4 100% 100% 96% 75% 63%M 5.5 100% 100% 88% 72% 56%M 5.6 96% 86% 90% 82% 71%UBH 1.2 87% 97% 54% 43% 39%UBH 1.3 93% 100% 80% 76% 64%UF 1.1 100% 100% 100% 93% 67%UF 1.2 100% 100% 100% 100% 89%UF 1.3 100% 91% 98% 84% 56%UF 1.4 100% 100% 97% 95% 85%UF 1.5 100% 100% 91% 77% 61%UF 2.4 100% 100% 99% 92% 90%UF 2.5 100% 100% 100% 90% 86%


Jack Pine Observed Climate Projected Climate

seedzones* 1961-1990 1997-2006 2020s 2050s 2080s

CM 1.1 99% 100% 89% 52% 31%CM 1.3 94% 87% 71% 32% 15%CM 2.1 100% 36% 69% 55% 22%CM 2.2 100% 10% 84% 71% 33%CM 3.1 97% 20% 72% 31% 14%CM 3.2 96% 16% 73% 37% 16%CM 3.3 88% 37% 67% 42% 21%DM 1.1 89% 41% 38% 10% 3%LBH 1.5 100% 54% 97% 70% 38%LF 1.1 81% 41% 86% 73% 47%NM 1.1 98% 91% 85% 69% 43%NM 2.1 93% 42% 57% 74% 42%UBH 1.2 78% 17% 23% 44% 34%


White spruce Observed Climate Projected Climate

seedzones* 1961-1990 1997-2006 2020s 2050s 2080s

BSA 1.1 98% 77% 96% 98% 90%BSA 1.2 99% 100% 100% 100% 91%CM 1.1 100% 100% 100% 98% 75%CM 1.2 100% 100% 100% 92% 67%CM 1.3 100% 100% 100% 98% 71%CM 2.1 100% 100% 100% 88% 79%CM 2.2 100% 100% 100% 92% 80%CM 2.3 100% 100% 100% 82% 69%CM 2.4 100% 99% 96% 85% 77%CM 3.1 99% 91% 85% 83% 77%CM 3.2 100% 97% 84% 81% 77%CM 3.3 100% 100% 98% 88% 74%CM 3.4 100% 100% 95% 95% 78%CM 3.5 100% 100% 100% 91% 71%DM 1.1 100% 100% 99% 85% 56%DM 1.2 99% 98% 88% 66% 50%DM 1.3 100% 100% 74% 74% 59%DM 2.1 73% 95% 74% 88% 57%DM 2.2 99% 99% 67% 87% 69%DM 2.3 100% 85% 87% 71% 58%LBH 1.1 100% 100% 100% 99% 76%LBH 1.2 98% 88% 100% 100% 86%LBH 1.3 100% 98% 100% 99% 82%LBH 1.4 100% 100% 100% 89% 69%LBH 1.5 100% 100% 100% 79% 81%LBH 1.6 100% 100% 99% 88% 69%LBH 2.1 100% 99% 100% 100% 86%LF 1.1 92% 100% 100% 98% 89%LF 1.2 100% 100% 95% 95% 74%LF 1.3 100% 99% 100% 100% 88%LF 1.4 100% 100% 100% 98% 80%LF 1.5 100% 100% 100% 98% 79%LF 2.1 100% 99% 100% 99% 80%LF 2.2 100% 100% 100% 100% 76%LF 2.3 100% 100% 78% 70% 50%M 1.1 100% 52% 25% 3% 1%M 2.1 100% 100% 92% 54% 33%M 2.2 100% 49% 69% 56% 30%M 3.2 100% 57% 80% 67% 41%M 4.1 100% 52% 83% 71% 42%M 4.2 100% 100% 100% 97% 68%M 4.3 98% 66% 54% 41% 30%M 5.1 98% 34% 74% 59% 36%M 5.3 98% 64% 85% 67% 44%M 5.4 47% 77% 95% 69% 51%NM 1.1 100% 100% 100% 100% 87%NM 2.1 99% 100% 100% 100% 85%UBH 1.1 99% 98% 99% 100% 82%UBH 1.2 100% 100% 100% 96% 80%UBH 1.3 94% 100% 100% 97% 86%UF 1.1 100% 100% 100% 93% 65%UF 1.2 100% 100% 100% 100% 89%UF 1.3 100% 83% 98% 79% 54%UF 1.4 100% 100% 100% 99% 77%UF 1.5 100% 100% 97% 87% 63%UF 2.4 100% 100% 100% 100% 80%UF 2.5 99% 96% 100% 94% 67%


Table S6. Table of best matching seed sources for 1961-1990 climate. The multivariate Mahalanobis climate distance is given in parenthesis.

Seed Zone Choice 1 Choice 2 Choice 3 Choice 4 Choice 5 Choice 6 Choice 7 Choice 8 Choice 9 Choice 10

Northern Mixedwood

NM11 NM11(0) KU11(0.5) LBH12(1.2) NM21(2.3) LBH21(2.6) PAD11(2.9) CM11(3.2)

NM21 NM21(0) LBH21(0.3) BSA12(0.8) LBH12(1) KU11(1.7) NM11(2.3) LBH11(3.2)

Central Mixedwood

CM11 CM11(0) PAD11(0) AP11(0.1) CM13(0.6) LBH11(0.8) CM12(1.7) DM11(2) LBH12(2.6) CM21(2.9) KU11(2.9)

CM12 CM12(0) CM21(0.4) DM11(0.4) CM22(1.2) UBH12(1.2) AP11(1.4) CM13(1.4) LBH14(1.4) CM11(1.7)

CM13 CM13(0) LBH11(0.5) AP11(0.6) CM11(0.6) PAD11(0.6) CM21(2) DM11(1) CM12(1.4) UBH12(2) LBH14(2.9)

CM21 CM21(0) CM12(0.4) CM22(0.5) DM11(0.6) LBH14(0.8) UBH12(1) CM23(1.2) CM24(1.4) LBH13(1.7) LBH15(1.7)

CM22 CM22(0) CM23(0.4) CM21(0.5) LBH14(0.5) CM24(1) LBH13(1) CM12(1.2) UBH12(1.2) DM11(1.7) LBH15(2)

CM23 CM23(0) CM22(0.4) CM24(0.8) LBH14(0.8) CM21(1.2) LBH13(1.7) CM12(2) UBH12(2) DM11(2.6) LBH15(2.6)

CM24 CM24(0) CM23(0.8) LBH15(0.8) LF11(0.8) CM22(1) CM21(1.4) CM31(1.7) LBH14(1.7)

CM31 CM31(0) CM33(0.5) LBH15(0.6) DM21(0.8) LF11(1) CM32(1.2) DM13(1.2) DM12(1.4) PRP11(1.4)

CM32 CM32(0) DM22(0.5) CM33(0.8) CM34(0.8) CP11(0.8) DM21(0.8) CM31(1.2) CP12(2) DM13(2) LF12(2)

CM33 CM33(0) CM31(0.5) DM13(0.6) LF11(0.6) CM32(0.8) CM34(0.8) DM21(0.8) PRP11(1.2) CP11(1.7) CP12(1.7)

CM34 CM34(0) CM32(0.8) CM33(0.8) CP11(1) CP12(1) DM22(1) DM21(1.2) LF12(1.2) DM13(1.4) DM23(1.7)

CM35 CM35(0) LF15(0.2) LF14(0.5) LF13(1) DM22(1.2) DM23(1.4) LF21(1.4) UF12(1.7) LF22(2)

Dry Mixedwood

DM11 DM11(0) CM12(0.4) CM21(0.6) CM13(1) LBH14(1.4) UBH12(1.4) AP11(1.7) CM22(1.7) CM11(2) LBH16(2)

DM12 DM12(0) DM13(0.8) LBH16(0.8) PRP11(0.8) CM31(1.4) LBH15(1.7) CM33(2) UBH13(2) CM24(2.3) DM21(2.3)

DM13 DM13(0) PRP11(0.3) CM33(0.6) DM12(0.8) DM21(1) CM31(1.2) CP12(1.2) CM34(1.4) CP11(1.7) LF12(1.7)

DM21 DM21(0) CP11(0.4) CM31(0.8) CM32(0.8) CM33(0.8) CP12(0.8) DM13(1) DM22(1) PRP11(1)

DM22 DM22(0) CP11(0.4) CM32(0.5) CM34(1) DM21(1) CM35(1.2) CP12(1.4) LF12(1.7) LF14(1.7) LF15(1.7)

DM23 DM23(0) LF21(0.5) LF22(0.5) LF15(0.8) LF14(1) CM35(1.4) UF14(1.4) CM34(1.7)

Boreal Highlands

BSA11 BSA11(0) BSA12(2.6) LBH12(2.6) LBH21(2.6)

BSA12 BSA12(0) LBH21(0.6) NM21(0.8) LBH12(2) BSA11(2.6) KU11(2.9)

LBH11 LBH11(0) CM13(0.5) AP11(0.8) CM11(0.8) PAD11(1) UBH12(2) CM12(2.3) DM11(2.3)

LBH12 LBH12(0) LBH21(0.6) KU11(1) NM21(1) NM11(1.2) BSA12(2)

LBH13 LBH13(0) UBH11(0.8) LBH14(1.2) UBH12(1.2) CM21(1.7) CM23(1.7) CM22(1) CM12(2.3) LBH11(3.2)

LBH14 LBH14(0) UBH12(0.4) CM22(0.5) CM21(0.8) CM23(0.8) LBH13(1.2) CM12(1.4) DM11(1.4) LBH16(1.4)

LBH15 LBH15(0) CM31(0.6) CM24(0.8) LF11(1) LBH16(1.2) CM21(1.7) CM33(1.7) DM12(1.7) CM22(2) LBH14(2)

LBH16 LBH16(0) DM12(0.8) UBH13(1) LBH15(1.2) LBH14(1.4) UBH12(1.4) CM31(2) DM11(2) CM21(2.3) CM24(2.3)

LBH21 LBH21(0) NM21(0.3) BSA12(0.6) LBH12(0.6) KU11(2) BSA11(2.6) NM11(2.6) LBH11(3.2)

UBH11 UBH11(0) LBH13(0.8) UBH12(1.7) LBH11(2.6) LBH14(2.6) CM21(3.2) CM22(3.2) CM12(3.6)

UBH12 UBH12(0) LBH14(0.4) CM21(1) CM12(1.2) CM22(1.2) LBH13(1.2) DM11(1.4) LBH16(1.4) UBH11(1.7)

UBH13 UBH13(0) LBH16(1) DM12(2) LBH15(2.3) PRP11(2.3) CM31(2.6) LF12(2.6) DM13(2.9) DM21(3.2)

Lower Foothills

LF11 LF11(0) CM33(0.6) CM24(0.8) CM31(1) LBH15(1) DM13(2) CM32(2.3) CM34(2.6) DM12(2.6)

LF12 LF12(0) CP11(1) CP12(1) CM34(1.2) DM21(1.2) PRP11(1.2) BWBSmw1(DM13(1.7) DM22(1.7) NF11(1.7)

LF13 LF13(0) CM35(1) UF12(1) LF14(1.2) LF15(1.4) BWBSwk1(2DM22(2.9) LF21(2.9) UF13(2.9)

LF14 LF14(0) LF15(0.4) CM35(0.5) LF21(0.6) UF12(0.6) DM23(1) LF13(1.2) LF22(1.4) UF14(1.4) DM22(1.7)

LF15 LF15(0) CM35(0.2) LF14(0.4) LF21(0.5) DM23(0.8) LF22(1) UF12(1.2) LF13(1.4)

LF21 LF21(0) LF22(0.3) DM23(0.5) LF15(0.5) LF14(0.6) UF14(0.8) UF12(1.2) CM35(1.4) UF13(2.3)

LF22 LF22(0) LF21(0.3) DM23(0.5) LF15(1) UF14(1.2) LF14(1.4) CM35(2) LF23(2.3) CM34(2.6) UF12(2.6)

LF23 LF23(0) UF15(0.3) FP11(1) M44(1.4) M43(1.7) LF22(2.3) M53(2.3) UF25(2.3) M32(2.6)

Montane

M11 M11(0) MG13(2.3)

M21 M21(0) UF13(0.8) UF14(2) UF24(2) SA11(2.6) UF12(2.6) LF14(3.6)

M22 M22(0) M32(0.5) FF11(1) M41(1.2) M45(1.2) FP11(1.4) M51(1.7) M55(1.7) MG11(1.7) LF12(2.3)

M32 M32(0) M22(0.5) M41(0.8) M45(1) UF25(1.2) FF11(1.4) FP11(1.4) M51(1.4) M55(1.4)

M41 M41(0) M32(0.8) M51(0.8) UF25(0.8) M53(1) M54(1) M22(1.2) M45(1.2) M55(1.4) M44(1.7)

M42 M42(0) SA31(1.2) UF25(2) M53(2.6) M54(2.9) UF15(2.9) M41(3.2) SA41(3.6)

M43 M43(0) M44(0.1) FP11(0.5) M53(0.5) M54(0.6) UF15(1) LF23(1.7) FF11(2) M41(2) UF25(2)

M44 M44(0) M43(0.1) FP11(0.5) M53(0.5) M54(0.5) UF15(0.8) LF23(1.4) M41(1.7) UF25(1.7) FF11(2)

M45 M45(0) M55(0.4) M51(1) M32(1) FF11(1.2) M22(1.2) M41(1.2) M54(1.2) FP12(1.4) M56(1.4)

M51 M51(0) M55(0.6) M41(0.8) M45(1) M54(1.2) M32(1.4) M53(1.4) SA32(1.4) M22(1.7) UF25(1.7)

M53 M53(0) M54(0.2) M43(0.5) M44(0.5) UF25(0.8) M41(1) UF15(1) FP11(1.2) M51(1.4) M45(1.7)

M54 M54(0) M53(0.2) M44(0.5) M43(0.6) M41(1) M45(1.2) M51(1.2) UF25(1.2) FP11(1.4) UF15(1.7)

M55 M55(0) M45(0.4) M51(0.6) M56(1) M32(1.4) M41(1.4) M22(1.7) MSdk(1.7) SA32(1.7)

M56 M56(0) SA33(0.8) M55(1) FP12(1.4) M45(1.4)

Upper Foothills

UF11 UF11(0)

UF12 UF12(0) LF14(0.6) UF13(0.8) LF13(1) LF15(1.2) LF21(1.2) UF14(1.4) CM35(1.7) DM23(2.3)

UF13 UF13(0) M21(0.8) UF12(0.8) UF14(1) UF24(1.4) LF14(1.7) LF21(2.3) SA11(2.3) DM23(2.9)

UF14 UF14(0) LF21(0.8) UF24(0.8) UF13(1) LF22(1.2) DM23(1.4) LF14(1.4) UF12(1.4) LF15(1.7)

UF15 UF15(0) LF23(0.3) FP11(0.8) M44(0.8) M43(1) M53(1) UF25(1.2) M54(1.7) M32(2)

UF24 UF24(0) UF14(0.8) SA11(1.4) UF13(1.4) UF25(1.4) M32(1.7) SA12(1.7) M21(2) M41(2) LF21(2.6)

UF25 UF25(0) M41(0.8) M53(0.8) M32(1.2) M54(1.2) UF15(1.2) SA31(1.4) UF24(1.4)

Table S7. Table of best matching seed sources for 1997-2006 climate. The multivariate Mahalanobis climate distance is given in parenthesis.


Northern Mixedwood

NM11 CM11(4.4) PAD11(4.4) AP11(4.8)

NM21 AP11(0.6) CM11(0.6) PAD11(0.8) LBH11(1.4) CM13(1.7) CM12(2) CM21(2.6) UBH12(2.6) LBH12(2.6)

Central Mixedwood

CM11 CM12(3.6) DM11(4) CM21(4.8)

CM12 CM12(3.2) CM24(3.2) CM21(3.6) CM23(3.6) DM11(4) CM31(4) LBH15(4) CM22(4) LBH16(4.4) DM12(4.4)

CM13 CM12(1.2) DM11(1.4) CM21(1.4) CM22(2) CM23(2) LBH14(2.3) CM24(2.3) LBH16(2.6) UBH12(2.6) LBH15(2.6)

CM21 CM31(2.6) LBH15(2.9) CM24(2.9) LF11(3.6) CM21(4) DM12(4) CM33(4) CM12(4.4) DM11(4.4) LBH16(4.4)

CM22 CM24(2) LBH15(2.6) CM23(2.6) CM31(2.9) LF11(2.9) CM21(3.2) CM22(3.2) CM12(3.6) LBH16(3.6) DM12(3.6)

CM23 CM24(2.6) CM23(2.6) CM22(3.6) LBH15(4) LF11(4) CM21(4.4) LBH14(4.4) DM12(4.4) CM12(4.8) LBH16(4.8)

CM24 LF11(2.3) CM31(2.6) CM24(2.6) CM33(2.9) LBH15(3.2) DM12(3.6) DM21(3.6) DM13(3.6) PRP11(4) CM32(4)

CM31 DM21(1.7) CM31(2.3) CM33(2.3) CM32(2.3) CP11(2.3) PRP11(2.9) DM13(2.9) CP12(2.9) DM22(2.9) LF11(3.2)

CM32 DM21(1.7) CM32(1.7) CP11(1.7) DM22(1.7) CM34(2) CM33(2.3) CP12(2.3) DM13(2.6) PRP11(2.9) LF12(2.9)

CM33 CM33(2) DM13(2) LF11(2.6) DM21(2.6) PRP11(2.6) CP12(2.6) CM34(2.6) CM31(2.9) CM32(2.9) NF11(2.9)

CM34 CM34(1.4) CP12(1.7) DM13(2.3) M22(2.3) M32(2.3) MG11(2.3) CM33(2.6) PRP11(2.6) LF12(2.6) NF11(2.6)

CM35 CM35(0.8) LF15(0.8) LF14(1) LF21(1.2) LF13(1.2) DM23(1.7) DM22(2) LF22(2) UF14(2.3) UF13(2.6)

Dry Mixedwood

DM11 CM24(2.3) CM23(2.6) CM12(2.9) DM11(3.2) CM21(3.2) LBH15(3.2) CM31(3.2) CM22(3.2) DM12(3.2) LBH16(3.6)

DM12 DM12(2.3) PRP11(2.6) DM13(2.6) CM31(3.6) CM33(3.6) NF11(3.6) LBH16(4) CM24(4) LF11(4) DM21(4)

DM13 PRP11(2) DM13(2) NF11(2.3) MG11(2.3) CP12(2.6) DM12(2.9) CM33(2.9) DMG11(2.9)M22(2.9)

DM21 CP11(1.4) DM21(1.7) CP12(2) DM22(2) CM32(2.3) NF11(2.3) CM34(2.6) CM33(2.9) PRP11(2.9) DM13(2.9)

DM22 DM22(1.2) CP11(1.4) CP12(2) CM34(2) DM21(2.3) CM32(2.3) LF14(2.3) CM35(2.3) LF12(2.6) LF15(2.6)

DM23 LF21(0.8) DM23(1) LF15(1) UF14(1.2) LF22(1.2) LF14(1.4) CM35(1.4) UF12(1.4) UF13(2) LF13(2.6)

Boreal Highlands

BSA11 BSA11(4.4) UBH11(4.8)

BSA12 LBH12(1.2) UBH12(4) CM11(2.3) KU11(2.3) PAD11(2.3) LBH11(2.6) LBH21(2.9) NM11(2.9) UBH11(3.2) BSA11(3.6)

LBH11 CM12(0.8) CM21(1) UBH12(1.2) LBH14(1.2) CM22(1.2) CM23(1.4) DM11(1.7) LBH13(2)

LBH12 CM11(2.9) PAD11(3.2) CM12(4) UBH12(4) CM13(4) LBH11(4.4) LBH12(4.4) DM11(4.8) CM21(4.8) NM11(4.8)

LBH13 CM23(2.3) CM22(2.6) CM21(2.9) UBH12(2.9) LBH14(2.9) CM24(2.9) CM12(3.2) LBH13(3.2) LBH15(3.6)

LBH14 CM23(1.7) CM24(2.3) LBH14(2.6) LBH16(2.9) CM22(2.9) CM21(3.2) UBH12(3.2) LBH15(3.2) CM12(3.6) DM12(3.6)

LBH15 CM31(1.7) LBH15(2.6) LF11(2.6) DM21(2.6) CM33(2.6) DM13(2.9) CM24(3.2) DM12(3.2) PRP11(3.2) CM32(3.2)

LBH16 DM12(2.3) LBH16(2.6) CM24(2.9) CM31(3.2) UBH13(3.2) LF11(3.2) DM13(3.2) LBH15(3.6) CM23(3.6) PRP11(3.6)

LBH21 CM11(1.2) PAD11(1.4) LBH11(1.7) CM12(2.3) UBH12(2.3) CM13(2.3) UBH11(2.3) LBH12(2.3) LBH13(2.9)

UBH11 UBH12(2.9) LBH14(3.6) CM12(4) CM21(4) LBH13(4) UBH11(4) LBH16(4.4) CM22(4.4) CM23(4.4) LBH15(4.8)

UBH12 CM23(2.3) UBH12(2.6) LBH14(2.6) CM22(2.9) CM21(3.2) LBH16(3.2) CM24(3.2) LBH13(3.2) CM12(3.6) LBH15(3.6)

UBH13 UBH13(2.6) DM12(2.9) LBH16(3.2) LF11(3.2) PRP11(3.2) M41(3.2) M51(3.2) DM13(3.6) M22(3.6) M32(3.6)

Lower Foothills

LF11 LF11(2) CM33(2.3) CM31(2.6) DM13(2.9) CM24(3.2) DM21(3.2) LBH15(3.6) PRP11(3.6) CM32(3.6) CM34(3.6)

LF12 PRP11(2) NF11(2) LF12(2) CP12(2) DM13(2.3) CM34(2.3) M32(2.3) MG11(2.3) CP11(2.9) FF11(2.9)

LF13 UF12(1) LF14(1.2) LF13(1.4) LF21(2) UF13(2) LF15(2) CM35(2.3) DM23(2.9) UF14(2.9) DM22(3.2)

LF14 LF21(0.8) LF14(1.2) UF12(1.2) DM23(1.4) UF14(1.4) UF13(1.4) LF22(1.7) M32(2) LF15(2) CM34(2.3)

LF15 LF21(0.6) LF15(0.6) LF14(0.8) CM35(1) UF12(1) DM23(1.2) LF22(1.4) LF13(1.4) UF14(1.7) UF13(2)

LF21 LF21(0.6) UF14(1) LF22(1.2) DM23(1.4) LF14(1.7) UF13(1.7) UF12(1.7) LF15(2) M32(2.3) UF24(2.3)

LF22 LF21(0.5) LF22(0.6) UF14(0.8) DM23(1.2) LF15(1.2) LF14(1.7) CM35(2) UF12(2) UF13(2.3) UF24(2.9)

LF23 LF23(1.2) UF15(1.7) UF14(2) LF22(2.3) UF24(2.9) LF21(2.9) FP11(2.9) DM23(3.2) UF25(3.2) M32(3.6)

Montane

M11 M11(4.4) M21(4.4)

M21 M21(0.6) UF24(2) UF13(2) UF14(2.6) SA11(2.9) M32(3.2) M22(4) M41(4) UF12(4)

M22 M22(0.5) M32(0.8) M41(1.7) FF11(1.7) M45(2) FP11(2) M55(2.3) UF24(2.3) M21(2.3) M51(2.6)

M32 M32(0.4) M22(0.8) UF24(1) M41(1.2) M21(1.4) UF25(1.7) UF14(1.7) M51(2) FP11(2) UF13(2)

M41 M41(1) UF24(1) M32(1.2) M22(1.7) M51(1.7) UF25(1.7) M21(2) M55(2.3) M45(2.3)

M42 UF24(1.4) SA12(2.3) UF14(2.3) UF25(2.6) SA11(2.9) UF13(2.9) M32(3.2) M41(3.6) M51(3.6)

M43 UF15(1.2) LF23(1.2) FP11(1.4) M41(1.7) UF25(1.7) M44(1.7) M32(2) M53(2) M43(2) UF24(2)

M44 UF24(1.7) UF15(1.7) LF23(1.7) UF14(2) UF25(2.3) M41(2.6) M32(2.6) M44(2.6) FP11(2.6) M53(2.9)

M45 UF24(2) M21(2.3) SA12(2.6) M41(2.6) M32(2.6) M55(2.6) M45(2.6) SA11(2.6)

M51 ESSFdc1(0.8) M41(1) M51(1) M32(1.2) UF24(1.2) SA12(1.4) M55(1.7) UF25(1.7) SA11(2) M45(2)

M53 UF24(0.8) UF25(1.4) UF14(1.4) M41(1.7) M32(1.7) UF15(2) M53(2.3) LF23(2.3) M21(2.6) M51(2.9)

M54 UF24(1) UF25(1.7) UF14(1.7) M41(2) M32(2.3) UF15(2.6) M21(2.6) SA12(2.9) M45(2.9) M53(2.9)

M55 UF24(2) SA12(2.3) SA11(2.3) M55(2.6) M21(2.6) M32(2.9) UF13(2.9) M41(3.2) M51(3.2) M45(3.2)

M56 ESSFmv2(2.3) ESSFwk2(2.6) SBSwk2(2.9) ESSFwc1(2.9) ESSFmm1(3.2)ESSFdc1(3.2) SA33(3.2) SA12(3.6) ESSFdk(3.6) ESSFwc4(3.6

Upper Foothills

UF11 UF11(1.4) LF13(4.4)

UF12 UF13(0.8) UF12(1.2) UF14(1.4) LF21(1.7) UF24(2) M21(2) LF14(2.3) LF22(2.9) DM23(3.2) LF15(3.2)

UF13 UF13(0.8) M21(1.2) UF24(1.4) UF14(1.7) SA11(1.7) UF12(2.3) M32(2.6) LF21(2.9) UF25(3.2) LF14(3.6)

UF14 UF14(0.5) UF24(1.2) UF13(1.2) M21(1.2) LF21(1.4) LF22(2) UF12(2) DM23(2.3) M32(2.6) LF14(2.6)

UF15 LF23(0.8) UF15(1) UF14(1.4) UF24(1.7) UF25(1.7) M32(2.3) LF22(2.3) FP11(2.3) M41(2.6) M53(2.6)

UF24 UF24(0.5) UF14(1.2) M21(1.2) UF13(1.4) SA11(1.7) M32(2.3) UF25(2.3) SA12(2.6) M41(2.9) UF12(2.9)

UF25 UF24(0.6) UF25(1) M41(2) UF14(2) UF15(2) M32(2.3) SA11(2.3) SA12(2.6) M53(2.6)

Table S8. Table of best matching seed sources for 2020s climate. The multivariate Mahalanobis climate distance is given in parenthesis


Northern Mixedwood

NM11 NM11(2.5) PAD11(3.9) CM11(4.2)

NM21 NM11(1.2) PAD11(1.3) CM11(1.5) CM13(1.7) LBH12(1.7) AP11(1.9) LBH11(2) KU11(2.1) NM21(2.6) LBH21(2.8)

Central Mixedwood

CM11 DM11(1.6) CM12(2.1) CM13(2.1) PAD11(2.2) CM11(2.3) AP11(2.4) CM21(3.3) LBH11(4.3)

CM12 DM11(1.5) CM12(2) CM21(2.3) CM31(2.4) LBH15(2.9) CM24(3.1) DM12(3.1) LBH16(3.2) CM22(3.8) LBH14(4.1)

CM13 DM11(0.8) CM13(1.8) CM12(1.9) CM21(2.5) LBH16(2.7) AP11(3.1) PAD11(3.1) CM11(3.2) LBH15(3.4)

CM21 CM31(1.3) DM11(1.8) LBH15(1.8) CM24(1.9) CM21(2.1) CM12(2.4) DM12(2.7) CM33(2.9) LBH16(3) LF11(3)

CM22 CM24(0.6) CM31(1.4) LBH15(1.4) CM21(1.7) CM23(1.7) LF11(1.8) CM22(1.9) DM12(2.1) CM33(2.3) DM11(2.3)

CM23 CM24(0.7) CM23(1.5) LF11(1.7) CM31(1.8) DM12(1.8) LBH15(1.9) CM22(2.2) CM33(2.2) LBH16(2.5)

CM24 CM33(0.8) CM31(1) LF11(1.2) CM24(1.5) DM13(1.6) CM32(1.7) DM21(2) LBH15(2.1) DM12(2.3) PRP11(2.6)

CM31 DM21(0.5) CM32(0.6) CP11(0.9) DM22(1.1) CM31(1.2) CM33(1.2) CM34(1.7) CP12(1.8) DM13(1.8) PRP11(2)

CM32 DM22(0.6) CM32(1) CP11(1.4) CM34(1.9) CM35(2) DM21(2) LF15(2.6) CM33(2.8) CP12(2.8) LF14(2.8)

CM33 CM32(0.6) CM34(0.7) DM22(0.9) CM33(1) CP11(1) DM21(1) CP12(1.6) DM13(1.6) LF12(2) CM31(2.1)

CM34 CM34(0.5) DM22(1.1) CP11(1.4) CM32(1.6) CP12(1.7) LF14(1.8) DM21(2.2) DM23(2.2) LF12(2.2) LF15(2.2)

CM35 LF13(1) CM35(1.3) LF15(2.1) LF14(2.7) DM22(3.4) UF12(3.5) LF21(4.4)

Dry Mixedwood

DM11 DM11(1.7) CM31(2.2) DM12(2.3) CM12(2.9) LBH16(2.9) LBH15(3) CM21(3.1) DM21(3.4) CM24(3.6) PRP11(3.7)

DM12 PRP11(0.4) DM13(0.6) DM12(1) DM21(1.3) CM33(1.8) CM31(1.9) CP12(1.9) NF11(2) LF12(2.1) CP11(2.3)

DM13 DM13(0.6) CM34(0.9) PRP11(0.9) DM21(1) CP12(1.1) CM33(1.2) CP11(1.2) LF12(1.7) CM32(1.8) NF11(2)

DM21 CP11(0.7) DM22(0.8) DM21(1.1) CM32(1.5) CP12(1.6) NF11(2.1) CM34(2.3) LF12(2.6) CM33(3) CM35(3.1)

DM22 DM22(1.1) CM35(1.5) CP11(2.1) LF13(2.1) LF15(2.4) CM32(2.6) LF14(2.7) DM21(3.5) CM34(3.6) CP12(3.7)

DM23 CM35(0.3) LF15(0.4) LF14(0.8) DM23(1.1) LF13(1.3) LF21(1.4) LF22(1.8) UF12(1.8) DM22(2) UF14(2.4)

Boreal Highlands

BSA11 LBH12(1.1) NM11(1.8) KU11(2.5) LBH21(3.2) PAD11(3.2) CM11(3.3) AP11(3.7) CM13(3.8) LBH11(3.8) NM21(4.2)

BSA12 NM11(0.7) LBH12(1.2) KU11(1.6) NM21(2) LBH21(2.1) PAD11(3.1) CM11(3.3) BSA12(3.7) LBH11(3.7) CM13(3.8)

LBH11 DM11(0.4) CM13(0.7) CM12(1.1) CM21(1.6) AP11(1.8) CM11(1.9) LBH11(2) PAD11(2) UBH12(2) LBH16(2.4)

LBH12 NM11(1.8) PAD11(2.3) CM11(2.5) CM13(2.5) LBH12(2.8) AP11(2.9) KU11(3.2) LBH11(3.7)

LBH13 CM21(0.4) CM22(0.5) LBH14(0.7) CM23(0.8) CM24(0.8) CM12(1.1) LBH15(1.1) DM11(1.2) UBH12(1.2) LBH16(1.5)

LBH14 DM12(0.7) LBH16(0.7) LBH15(1.1) CM24(1.3) CM31(1.3) DM11(1.6) LBH14(1.7) CM21(1.9) CM23(2.1) CM22(2.3)

LBH15 CM31(0.2) CM33(0.6) DM21(0.6) CM32(0.7) DM13(1.3) LBH15(1.5) CP11(1.6) LF11(1.7) PRP11(1.7) DM12(1.9)

LBH16 DM12(0.6) PRP11(0.8) DM13(1.2) LBH16(1.3) UBH13(1.3) CM31(1.5) DM21(1.8) CM33(2.1) LF12(2.1) LBH15(2.6)

LBH21 NM11(1.4) LBH12(1.5) CM13(1.7) PAD11(1.7) CM11(1.8) LBH11(1.9) AP11(2.2) KU11(2.4) LBH21(2.7) NM21(2.8)

UBH11 UBH12(0.4) CM21(0.7) LBH14(0.8) CM12(1) DM11(1.1) LBH16(1.2) CM22(1.4) LBH15(1.6) CM13(1.9) CM23(2.1)

UBH12 LBH16(0.5) LBH15(1.1) DM11(1.2) DM12(1.3) UBH13(1.5) CM21(1.6) CM31(1.6) LBH14(1.7) UBH12(1.7) CM12(1.9)

UBH13 LF12(0.9) UBH13(1.1) PRP11(1.5) DM21(2) DM12(2.4) DM13(2.4) CP12(2.5) NF11(2.5) CM31(2.6) CP11(2.6)

Lower Foothills

LF11 CM33(0.4) CM32(0.6) CM31(1.4) LF11(1.4) DM21(1.5) DM13(1.6) DM22(1.8) CP11(2) PRP11(2.6) CP12(2.7)

LF12 LF12(0.7) CP11(1) CP12(1.1) CM34(1.4) DM22(1.5) DM21(1.8) NF11(1.8) PRP11(2.3) LF14(2.4) CM32(2.7)

LF13 LF13(0.8) CM35(2.5) UF12(2.9) LF14(3.2) LF15(3.2)

LF14 LF14(0.5) LF13(0.7) CM35(0.9) LF15(0.9) UF12(1) LF21(1.8) DM23(2.3) DM22(2.6) UF13(2.7) LF22(2.9)

LF15 CM35(0.7) LF13(0.7) LF15(1.1) LF14(1.6) UF12(2.1) LF21(2.6) DM22(3.1) DM23(3.2) LF22(3.5) UF14(3.9)

LF21 LF15(0.5) LF14(0.7) LF21(0.8) UF12(0.8) CM35(0.9) LF13(1.1) LF22(1.6) DM23(1.8) UF14(1.8) UF13(2.4)

LF22 LF15(0.3) CM35(0.6) LF21(0.8) LF14(1) LF22(1) DM23(1.3) LF13(1.8) UF12(1.8) UF14(1.9) DM22(2.9)

LF23 LF22(0.6) LF23(0.9) DM23(1.2) LF21(1.2) UF14(1.5) UF15(1.5) FP11(2.1) M32(2.4) LF15(2.6) LF14(2.9)

Montane

M11 M11(0.7) MG13(2.3)

M21 M21(0.5) UF13(1) UF12(2.5) UF14(2.5) UF24(3.3) LF14(3.7) LF13(4.2) SA11(4.2) LF21(4.4)

M22 M22(1.1) FF11(1.2) MG11(1.2) LF12(1.5) M32(1.5) CP12(1.6) M55(1.6) M45(1.7) NF11(2) MG12(2.2)

M32 M32(0.9) CP11(3) DM23(1.3) LF14(1.3) UF14(1.4) M22(1.6) M45(1.6) UF13(1.6) LF12(1.7) LF21(1.7)

M41 M32(0.5) M45(0.7) M55(0.8) M41(1) M22(1.2) UF24(1.2) M51(1.5) FP12(1.6) FF11(1.8)

M42 UF24(1.4) UF14(1.8) FP12(2.2) UF25(2.5) M42(2.8) SA12(2.9) M32(3.2) M41(3.2) LF21(3.5) LF22(3.5)

M43 FP11(0.5) M44(0.5) M43(0.7) FF11(1) M54(1) UF15(1) M32(1.1) M45(1.1) M53(1.1) LF23(1.3)

M44 FP11(0.8) M44(0.9) FF11(1.1) M45(1.1) M32(1.2) UF15(1.3) FP12(1.4) ICHmk3(1.4)M43(1.4) M54(1.4)

M45 M56(1) FP12(1.1) M45(1.2) M55(1.2)

M51 M45(0.7) M55(0.8) FP12(1) M32(1) M56(1.1) M51(1.2) M41(1.3) UF24(1.8) M22(1.9)

M53 M45(0.7) M32(0.8) FP12(0.9) M55(1.2) M54(1.3) M53(1.4) M51(1.5) FF11(1.6) M41(1.6) M44(1.6)

M54 M45(0.4) FP12(0.8) M55(0.8) M32(1) FF11(1.4) M54(1.4) M51(1.5) M56(1.5) M41(1.6) M53(1.7)

M55 M56(0.9) M55(1.1) FP12(1.6) M45(1.6) SA33(2.3) M32(2.7) FF11(3) LF12(3.1) M51(3.1) M22(3.3)

M56 M56(1.2) FP12(1.9) ICHmk1(1.9) MSdk(1.9) SA33(2) SBSwk2(2.1) ICHwk1(2.2) 17x(2.3) ESSFwc4(2.3)SBSvk(2.4)

Upper Foothills

UF11 UF11(1.1)

UF12 LF13(0.7) UF12(0.9) LF14(1.9) CM35(2.2) LF15(2.2) UF13(2.3) LF21(3.1) UF14(3.4) M21(4.3)

UF13 UF13(0.7) UF12(0.8) LF13(1.6) M21(1.8) LF14(2) UF14(2.4) LF15(3) CM35(3.1) LF21(3.2) UF24(3.9)

UF14 UF12(0.8) UF14(1.1) LF13(1.3) LF14(1.3) UF13(1.3) LF15(1.4) CM35(1.5) LF21(1.7) DM23(2.4) M21(2.4)

UF15 LF22(0.7) LF23(0.9) UF15(1.1) DM23(1.2) LF21(1.2) UF14(1.2) M32(1.5) FP11(1.7) M44(2.2) UF24(2.2)

UF24 UF13(0.6) UF14(0.9) UF12(1) M21(1.3) LF14(1.6) UF24(1.8) LF13(2) LF15(2.3) LF21(2.3) CM35(2.4)

UF25 UF24(0.6) UF14(0.7) M32(1.1) UF25(1.5) LF21(1.7) UF13(1.7) FP12(1.8) LF22(1.9) DM23(2.1) M41(2.1)

Table S9. Table of best matching seed sources for 2050s climate. The multivariate Mahalanobis climate distance is given in parenthesis.


Northern Mixedwood

NM11 BWBSmw2(3.8)

NM21 CM13(2.1) PAD11(2.5) CM11(2.7) AP11(3.1) DM11(3.1) LBH11(3.8) CM12(4.3) NM11(4.3)

Central Mixedwood

CM11 DM11(3.9) CM31(4.8)

CM12 CM31(3) DM21(3.3) DM12(4.1) CM32(4.3) CM33(4.5) PRP11(4.6) DM13(4.7) DM11(4.9)

CM13 CM31(3) DM11(3) DM12(3.2) DM21(3.4) LBH16(3.8) PRP11(4.1) LBH15(4.4) CM12(4.5) UBH13(4.6) DM13(4.8)

CM21 CM31(2.2) DM21(2.5) CM32(2.8) CM33(3.1) DM13(3.8) DM12(3.9) CP11(4) CM24(4.1) DM22(4.1) LBH15(4.1)

CM22 CM31(1.6) CM33(1.9) CM32(2.2) DM21(2.3) DM13(2.5) CM24(2.6) LF11(2.8) DM12(2.9) LBH15(3.2) PRP11(3.3)

CM23 CM33(1.8) CM31(2) DM13(2) DM21(2.3) CM32(2.6) DM12(2.6) PRP11(2.6) LF11(2.9) CM24(3)

CM24 CM32(1.5) CM33(1.7) DM21(1.9) DM13(2.2) CM31(2.4) DM22(2.4) CM34(2.5) CP11(2.5) PRP11(3.1) LF11(3.3)

CM31 DM22(1.2) CP11(1.4) DM21(1.8) CM32(1.9) CP12(2.6) CM34(2.8) NF11(3.2) CM33(3.3) CM35(3.6) DM13(3.6)

CM32 DM22(1.8) CM35(2.5) CM32(3) CP11(3) LF13(3.3) LF15(3.4) CM34(3.9) LF14(3.9) DM21(4.2) CP12(4.5)

CM33 DM22(1) CP11(1.6) CM32(1.7) CM34(1.8) DM21(2.3) CP12(2.5) CM35(2.6) CM33(3) LF14(3.1) LF15(3.1)

CM34 DM22(1.6) CM35(1.8) CM34(2) LF15(2.1) LF14(2.2) CP11(2.4) CM32(2.9) DM23(2.9) CP12(3) LF13(3)

CM35 CM35(3.8) LF13(3.3)

Dry Mixedwood

DM11 DM21(2.6) CM31(2.9) DM12(3.4) PRP11(3.5) DM13(3.9) CP11(4.1) CM32(4.2) CM33(4.3) NF11(4.9)

DM12 PRP11(1.2) DM21(1.4) CP11(1.5) DM13(1.5) CP12(1.6) NF11(1.7) LF12(2.2) CM34(2.5) DM22(2.6) CM33(2.7)

DM13 CP11(1.3) CP12(1.5) DM22(1.5) CM34(1.6) DM21(1.9) DM13(2.2) CM32(2.4) PRP11(2.4) NF11(2.5) LF12(2.7)

DM21 DM22(2) CP11(2.3) CM35(3.4) CP12(3.4) DM21(3.4) NF11(3.7) CM32(3.8) LF13(4.2) CM34(4.5) LF15(4.7)

DM22 CM35(2.9) CP11(4.6) DM22(3.3) LF13(3.2) LF14(4.8) LF15(4.2)

DM23 CM35(1.3) LF13(1.7) LF15(1.9) LF14(2.7) DM23(3.3) LF21(3.5) UF12(3.7) DM22(3.8) LF22(4) UF14(4.7)

Boreal Highlands

BSA11 BWBSdk2(2.5) BWBSmw2(3) NM11(4.7) CM13(4.8)

BSA12 NM11(2.8) CM13(3.6) PAD11(3.7) CM11(4) LBH12(4.3) AP11(4.6) KU11(4.9)

LBH11 DM11(1.6) LBH16(2.4) DM12(2.6) CM12(2.8) CM31(2.8) CM21(3.2) LBH15(3.3) CM13(3.6) UBH13(3.6)

LBH12 BWBSmw2(2.5 BWBSdk2(3.3)

LBH13 CM31(0.9) DM12(1.5) CM24(1.6) LBH15(1.7) CM33(1.8) LBH16(2.1) LF11(2.1) DM13(2.3) DM21(2.3) CM32(2.7)

LBH14 DM12(1.2) CM31(1.3) PRP11(1.3) DM13(1.4) DM21(1.4) CM33(1.9) LBH16(2.5) CM32(2.7) CP11(2.7) LBH15(2.9)

LBH15 DM21(0.7) CM32(0.8) DM22(0.9) CP11(1) CM33(1.4) CM31(1.6) CM34(1.8) DM13(1.9) CP12(2) PRP11(2.1)

LBH16 PRP11(0.9) DM21(1.2) DM13(1.5) LF12(1.6) CP11(1.8) CP12(2) DM12(2) NF11(2) CM31(2.5) CM33(2.5)

LBH21 CM13(2.4) PAD11(3.3) CM11(3.5) DM11(3.6) AP11(3.8) LBH11(4.2) NM11(4.7)

UBH11 CM31(1.1) LBH16(1.1) LBH15(1.4) DM12(1.6) UBH13(1.8) DM11(2) CM24(2.4) CM21(2.5) DM21(2.6) CM33(2.7)

UBH12 CM31(1.4) DM12(1.4) DM21(1.7) LBH16(1.9) PRP11(1.9) UBH13(1.9) CP12(4) DM13(2.3) CM33(2.6) LBH15(2.6)

UBH13 LF12(1.2) NF11(2) CP11(2.1) CP12(2.1) PRP11(2.1) DM21(2.2) DM13(3.2) DM22(3.2) CM34(3.4) UBH13(3.4)

Lower Foothills

LF11 CM32(1.1) CM34(1.3) DM22(1.3) CM33(1.7) CP11(2.1) DM21(2.2) DM13(2.8) CM35(3.1) CP12(3.1) CM31(3.2)

LF12 CP11(1.5) DM22(1.6) CP12(1.7) LF12(2) CM34(2.2) LF14(2.3) NF11(2.4) CM35(2.7) DM21(2.8) LF13(3)

LF13 LF13(3.1) CM35(4.8)

LF14 LF13(1.2) CM35(1.6) LF15(2) LF14(2.1) UF12(2.8) LF21(3.5) DM23(3.8) DM22(4.1) LF22(4.4) UF13(4.9)

LF15 LF13(2.6) CM35(3) LF15(3.9) LF14(4.8)

LF21 LF13(1.4) CM35(1.8) LF15(1.9) LF14(2.4) UF12(2.5) LF21(2.9) LF22(3.8) DM23(3.9) UF14(4) UF13(4.5)

LF22 CM35(1.5) LF15(1.7) LF13(1.8) LF14(2.6) LF21(3) UF12(3.2) LF22(3.4) DM23(3.5) UF14(4.1) DM22(4.4)

LF23 LF22(0.8) DM23(1) LF15(1) LF21(1) CM35(1.6) LF14(1.6) UF14(1.6) UF12(2.5) LF23(2.7) LF13(3)

Montane

M11 43x(2) M11(2.2) 17aj(2.3) 17al(2.3) 80b(2.5) 80c(2.5) 43w(2.6) 17ab(2.7) 18d(2.8) 18b(2.9)

M21 M21(1.8) UF13(2.3) UF12(3.1) LF13(3.2) UF14(4.3) LF14(4.4) UF11(4.4)

M22 FF11(1.6) MG11(1.6) LF14(1.8) CP12(1.9) LF12(1.9) M22(1.9) M45(1.9) DM23(2.1) M32(2.1) M55(2.2)

M32 LF14(0.8) UF12(1.4) UF13(1.4) CM35(1.6) DM23(1.6) LF15(1.6) LF13(1.7) UF14(1.7) LF21(1.8) FP12(1.9)

M41 FP12(1.8) M55(1.8) UF13(1.8) LF14(1.9) M45(1.9) M56(2) M32(2.1) UF14(2.1) M21(2.3) UF24(2.3)

M42 UF14(1.7) LF14(1.8) FP12(1.9) CM35(2.1) LF15(2.2) UF12(2.2) LF13(2.3) LF21(2.4) UF13(2.5) UF24(2.7)

M43 FF11(1) FP12(1) DM23(1.3) M45(1.3) M32(1.5) FP11(1.7) LF22(1.9) M22(2) M44(2)

M44 FP12(1) DM23(1.4) FF11(1.5) M45(1.7) LF22(1.9) M32(1.9) FP11(2.1) LF21(2.1) UF14(2.1)

M45 FP12(1.6) M56(2.1) M45(2.6) M55(2.8) LF14(3.2) FF11(3.4) CM35(3.7) LF12(3.7) MG11(3.8) CP12(3.9)

M51 M56(1.1) FP12(1.4) M55(1.7) M45(1.9) LF14(2.5) LF12(2.6) M32(2.6) SA33(2.7) UF13(2.7)

M53 FP12(0.9) M45(1.7) LF14(1.8) M56(1.8) DM23(1.9) M55(2) M32(2.1) UF14(2.2) FF11(2.3)

M54 FP12(0.7) M45(1.5) M56(1.7) M55(1.9) LF14(2.2) M32(2.2) DM23(2.3) FF11(2.3) UF14(2.3) CM35(2.7)

M55 FP12(1.9) M56(2.2) M45(3) M55(3) UF13(3.5) CM35(3.8) LF12(3.9) LF13(3.9) SA33(3.9) UF14(3.9)

M56 ICHmw2(1.9) ESSFwc1(2.1) ESSFwk2(2.8)ICHdw(2.9) SBSwk2(3) 15o(3.2) BWBSwk1(3.2 ICHwk1(3.2) 17x(3.3) 80c(3.3)

Upper Foothills

UF11 UF11(3.8)

UF12 LF13(2.3) UF12(3.7) UF11(4.2) CM35(4.4) LF15(4.8) LF14(4.9)

UF13 LF13(1.8) UF12(1.9) UF13(2.3) LF14(3.1) M21(3.2) CM35(3.6) LF15(3.8) UF14(3.8) UF11(4.2) LF21(4.4)

UF14 LF13(1.4) UF12(2.1) CM35(2.4) LF15(2.7) LF14(2.8) UF13(2.9) UF14(3) LF21(3.6) M21(4) DM23(4.5)

UF15 LF21(0.7) LF22(0.7) DM23(0.8) LF15(0.9) UF14(1) LF14(1.1) CM35(1.4) UF12(1.9) FP12(2.3) LF23(2.5)

UF24 LF13(1.5) UF13(1.8) UF12(1.9) M21(2.5) UF14(2.7) LF14(2.8) CM35(2.9) LF15(3.3) UF24(4)

UF25 UF14(0.9) LF14(1.2) UF12(1.2) UF13(1.2) LF13(1.7) LF15(1.7) LF21(1.7) CM35(1.8) UF24(1.8) DM23(2.1)

Table S10. Table of best matching seed sources for 2080s climate. The multivariate Mahalanobis climate distance is given in parenthesis.


Northern Mixedwood

NM11 BWBSmw2(5.9)BWBSdk2(7) BWBSmw1(8.9)

NM21 DM11(3.8) CM13(4.8) CM12(5.4)

Central Mixedwood

CM11 42i(5.2)

CM12 42i(3.8) 42k(4.2) DM21(4.3) CM32(5.2) CP11(5.2) CM31(5.3)

CM13 DM21(2.9) CM31(3.7) PRP11(3.8) CP11(4) DM12(4) DM13(4.5) NF11(4.5) CM32(4.8) DM22(4.9)

CM21 DM21(3.7) CM32(3.8) DM22(4.2) CP11(4.4) CM31(4.6) CM33(4.9) DM13(5.2)

CM22 CM32(2.8) DM21(3) CM33(3.4) DM22(3.4) CP11(3.6) DM13(3.6) CM31(3.7) PRP11(4.3) CM34(4.5) CP12(5)

CM23 DM21(2.8) CM32(2.9) DM13(3.1) CM33(3.2) CP11(3.2) DM22(3.3) PRP11(3.7) CM31(3.9) CM34(3.9)

CM24 DM22(2.6) CM32(2.7) CP11(3.1) DM21(3.3) CM34(3.5) CM33(3.7) DM13(3.9) CP12(4.2) PRP11(4.7)

CM31 DM22(2.4) CP11(2.8) CP12(3.7) DM21(3.7) CM32(3.8) CM35(3.9) CM34(4.2) NF11(4.3) LF13(4.8) LF15(5)

CM32 CM35(3.7) DM22(3.7) LF13(4.6) LF15(4.6) CP11(5) CM32(5.2)

CM33 DM22(2.2) CM35(2.8) CP11(3.2) CM32(3.5) LF15(3.5) CM34(3.6) LF13(3.8) LF14(4) CP12(4.1) DM21(4.4)

CM34 DM22(3) CM35(2.5) DM23(4) LF15(3) LF14(3.5) CM34(3.7) LF13(3.7) CP11(4.1) CP12(4.6) LF21(4.6)

CM35 17b(3.6) 17a(5.1)

Dry Mixedwood

DM11 42i(2.8) 42k(2.9) DM21(3.4) CP11(4) DM22(4.6) NF11(4.6) PRP11(4.6) CM32(4.8) CM31(4.9) DM13(4.9)

DM12 CP11(1.7) CP12(1.9) NF11(2) DM22(2.3) DM21(2.4) PRP11(2.7) DMG11(2.9) DM13(3) CM34(3.1) LF12(3.3)

DM13 DM22(2.2) CP11(2.3) CP12(2.7) CM34(2.9) DM21(3.6) CM32(3.7) CM35(3.7) NF11(3.7) DM13(4.1) MG11(4.3)

DM21 DM22(3.7) CP11(4) CM35(4.4) CP12(4.8) LF13(4.9) NF11(5.1)

DM22 CM35(4.3) LF13(4.5) DM22(5.2)

DM23 CM35(3) LF13(3.3) LF15(3.7) LF14(4.7) DM23(5.2) LF21(5.3)

Boreal Highlands

BSA11 BWBSdk2(2.7) BWBSmw2(2.8)

BSA12 BWBSmw2(2.2)BWBSdk2(3.8) CM13(5.1)

LBH11 DM21(2.6) DM12(2.8) CM31(2.9) PRP11(3.3) DM13(3.9) LBH16(4) CP11(4.1) CM32(4.6) CM33(4.6) UBH13(4.7)

LBH12 BWBSmw2(3.6)BWBSdk2(4.6)

LBH13 DM21(1.7) CM31(1.9) CM32(2) CM33(2) DM13(2.3) PRP11(2.6) CP11(2.7) DM22(2.8) DM12(2.9) LF11(3.4)

LBH14 DM21(1.5) CP11(1.9) PRP11(2) DM13(2.2) DM22(2.5) CM32(2.7) CP12(2.7) NF11(2.7) CM31(2.9) CM33(2.9)

LBH15 DM22(1.3) CP11(1.5) CM32(2) DM21(2) CP12(2.5) CM34(2.7) CM35(3.1) CM33(3.2) NF11(3.3) DM13(3.4)

LBH16 CP11(1.5) DM21(1.7) CP12(1.8) NF11(1.8) PRP11(1.9) LF12(2.2) DM22(2.3) DM13(2.4) CM34(2.9) DMG11(3.1)

LBH21 BWBSmw2(2.9)BWBSdk2(4.3) DM11(4.5) LBH16(5.2) UBH13(5.2) CM13(5.3)

UBH11 DM21(1.2) CM31(1.4) PRP11(2) CM32(2.1) DM12(2.1) CM33(2.2) CP11(2.2) DM13(2.2) DM22(2.6) LF12(2.8)

UBH12 DM21(1.5) CP11(1.9) PRP11(2.2) DM22(2.6) LF12(2.7) NF11(2.7) CM31(2.8) DM13(2.8) CP12(2.9) CM32(3)

UBH13 LF12(1.8) CP11(2) NF11(2) CP12(2.1) DM22(2.8) DM21(2.9) PRP11(3.2) CM34(3.5) MG11(3.6) DMG11(3.7)

Lower Foothills

LF11 DM22(2) CM32(2.6) CM34(2.7) CM35(2.9) CP11(3.1) LF15(3.3) LF14(3.7) CM33(3.9) LF13(3.9) DM21(4)

LF12 DM22(2.2) CM35(2.4) CP11(2.5) CP12(2.7) LF13(2.7) LF14(2.7) LF15(3) CM34(3.2) DM23(3.5) LF12(3.6)

LF13 17b(5.9) LF13(6.1) CM35(7.2) UF11(7.2) LF15(8)

LF14 LF13(2.9) CM35(3) LF15(3.4) LF14(3.9) LF21(4.8) UF12(4.8) DM23(5.1)

LF15 17b(4) LF13(5) CM35(5.4)

LF21 LF13(3.3) CM35(3.6) LF15(3.9) LF14(4.8) UF12(5.1) LF21(5.2)

LF22 CM35(3.3) LF13(3.6) LF15(3.8) LF14(4.9) LF21(5.3)

LF23 LF15(1.7) CM35(1.8) LF22(2.2) DM23(2.3) LF21(2.3) LF14(2.7) LF13(2.9) UF14(3.1) UF12(3.7)

Montane

M11 43w(1.7) 43x(2) 80c(2.7) 80b(2.9) 43o(3.1) 17a(3.2) 17aj(3.2) 18b(3.2) 17al(3.3) 43q(3.3)

M21 17t(1.9) M21(3.4) UF11(4) UF13(4.3) LF13(4.4) UF12(4.9)

M22 IDFmw1(1.9) IDFmw2(2) 15c(2.2) PPdh2(2.2) 42q(2.3) BWBSwk1(2.3)IDFdm2(2.4) MG11(2.4) LF14(2.5) CM35(2.6)

M32 LF13(1.3) LF14(1.6) CM35(1.7) LF15(2.1) UF12(2.2) DM23(2.6) UF13(2.6) FP12(2.7) LF21(2.9) UF14(2.9)

M41 BWBSwk1(0.9) LF13(1.9) UF13(2.3) LF14(2.4) CM35(2.6) FP12(2.6) UF12(2.6) M21(2.7) UF14(2.8)

M42 LF13(2.2) CM35(2.7) LF14(3.1) LF15(3.4) UF12(3.4) FP12(3.5) UF14(3.5) UF13(3.8) LF21(4.4) M56(4.5)

M43 FP12(1.5) CM35(1.6) DM23(1.6) LF15(1.8) LF14(1.9) FF11(2.2) LF21(2.3) LF22(2.3) M45(2.5) UF14(2.7)

M44 FP12(1.7) CM35(1.8) DM23(1.8) LF15(1.9) LF14(2.1) LF21(2.4) LF22(2.4) UF14(2.7) FF11(2.8) LF13(3)

M45 43v(2.1) 17a(2.4) ICHmw2(2.6) BWBSwk1(2.8)FP12(2.8) ICHdw(2.9) 15c(3) 43q(3.1) ICHmk1(3.1) IDFmw1(3.3)

M51 BWBSwk1(1.1) SBSwk2(1.8) ESSFmv2(1.9) FP12(1.9) ICHmw2(1.9) ICHmk1(2.1) M56(2.1) ESSFwc1(2.2)ICHmm(2.2) MSdk(2.4)

M53 FP12(1.4) CM35(1.7) LF14(2) LF15(2.1) LF13(2.3) DM23(2.5) UF14(2.7) M56(2.8) LF21(2.9) M45(2.9)

M54 FP12(1.7) CM35(2.2) LF14(2.5) LF13(2.7) LF15(2.7) DM23(3) M56(3) M45(3.1) UF14(3.2) LF21(3.5)

M55 17a(2.1) BWBSwk1(2.3) 43v(2.7) 17b(3.1) 17t(3.2) ICHmw2(3.2) ESSFmv2(3.3)43q(3.4) 43d(3.5) FP12(3.5)

M56 17a(3.6) ICHmw2(3.6) 15o(3.9) BWBSwk1(4.1)ESSFwc1(4.1)80c(4.2) ICHdw(4.4) ESSFmv2(4.5)17b(4.6)

Upper Foothills

UF11 UF11(8.3) 17b(12.1)

UF12 LF13(5.3) UF11(5.4)

UF13 LF13(3.2) UF11(3.6) UF12(4.2) CM35(4.8) UF13(4.8) LF14(4.9) LF15(5.1) M21(5.2)

UF14 LF13(3) UF11(3.9) CM35(4.1) UF12(4.4) LF15(4.6) LF14(4.9) UF14(5.1) UF13(5.3)

UF15 CM35(1.1) LF15(1.2) LF14(1.7) LF21(1.7) DM23(1.8) LF13(1.8) LF22(1.9) UF14(2.1) UF12(2.5) FP12(3.1)

UF24 UF11(2.3) LF13(2.5) UF12(3.6) UF13(3.7) CM35(4.2) M21(4.3) LF14(4.5) UF14(4.5) LF15(4.8)

UF25 LF13(1.4) CM35(1.9) LF14(2.1) UF12(2.1) LF15(2.3) UF14(2.3) UF13(2.5) LF21(3) DM23(3.3) FP12(3.3)

Recommended seed choice Zone/Ecoregion

BWBSdk2, BWBSmw1, BWBSmw2, BWBSwk1 Boreal White and Black Spruce zoneESSFmv2, ESSFwc1 ESSFwk2 Engelmann Spruce-Subalpine Fir zoneICHdw, ICHmm, ICHmk1, ICHmw2, ICHwk1 Interior Cedar-Hemlock zoneIDFdm2, IDFmw1, IDFmw2 Interior Douglas-fir zoneMSdk Montane Spruce zonePPdh2 Ponderosa Pine zoneSBSwk2 Sub-Boreal Spruce zone

15o Northern Rockies ecoregion17ab Middle Rockies ecoregion80b, 80c Northern Basin and Range ecoregion

15c Northern Rockies ecoregion17al, 17aj, 17t, 17x Middle Rockies ecoregion42i, 42k, 42q Northwestern Glaciated Plains ecoregion43d, 43o, 43v Northwestern Great Plains ecoregion

17b Middle Rockies ecoregion

17a Middle Rockies ecoregion18b, 18d Wyoming Basin ecoregion43q, 43x, 43w Northwestern Great Plains ecoregion

Wyoming

Table S11. Locations of recommended seed choices which originate outside of Alberta. For British Columbia we report the relevant ecological "variants" and "zones" [18], and for the United States we report the corresponding state and "level III & IV" ecoregions [20].

British Columbia

Idaho

South Dakota

Montana

Documents

Strategies for Reforestation under Uncertain Future …ahamann/publications/pdfs/...Strategies for Reforestation under Uncertain Future Climates: Guidelines for Alberta, Canada Laura